Multi-Task Learning for Calorie Prediction on a Novel Large-Scale Recipe Dataset Enriched with Nutritional Information

Robin Ruede, Verena Heusser, Lukas Frank, Monica Haurilet, Alina Roitberg, Rainer Stiefelhagen

Responsive image

Auto-TLDR; Pic2kcal: Learning Food Recipes from Images for Calorie Estimation

Slides Poster

A rapidly growing amount of content posted online, such as food recipes, opens doors to new exciting applications at the intersection of vision and language. In this work, we aim to estimate the calorie amount of a meal directly from an image by learning from recipes people have published on the Internet, thus skipping time-consuming manual data annotation. Since there are few large-scale publicly available datasets captured in unconstrained environments, we propose the pic2kcal benchmark comprising 308,000 images from over 70,000 recipes including photographs, ingredients and instructions. To obtain nutritional information of the ingredients and automatically determine the ground-truth calorie value, we match the items in the recipes with structured information from a food item database. We evaluate various neural networks for regression of the calorie quantity and extend them with the multi-task paradigm. Our learning procedure combines the calorie estimation with prediction of proteins, carbohydrates, and fat amounts as well as a multi-label ingredient classification. Our experiments demonstrate clear benefits of multi-task learning for calorie estimation, surpassing the single-task calorie regression by 9.9%. To encourage further research on this task, we make the code for generating the dataset and the models publicly available.

Similar papers

Picture-To-Amount (PITA): Predicting Relative Ingredient Amounts from Food Images

Jiatong Li, Fangda Han, Ricardo Guerrero, Vladimir Pavlovic

Responsive image

Auto-TLDR; PITA: A Deep Learning Architecture for Predicting the Relative Amount of Ingredients from Food Images

Slides Poster Similar

Increased awareness of the impact of food consumption on health and lifestyle today has given rise to novel data-driven food analysis systems. Although these systems may recognize the ingredients, a detailed analysis of their amounts in the meal, which is paramount for estimating the correct nutrition, is usually ignored. In this paper, we study the novel and challenging problem of predicting the relative amount of each ingredient from a food image. We propose PITA, the Picture-to-Amount deep learning architecture to solve the problem. More specifically, we predict the ingredient amounts using a domain-driven Wasserstein loss from image-to-recipe cross-modal embeddings learned to align the two views of food data. Experiments on a dataset of recipes collected from the Internet show the model generates promising results and improves the baselines on this challenging task.

RWMF: A Real-World Multimodal Foodlog Database

Pengfei Zhou, Cong Bai, Kaining Ying, Jie Xia, Lixin Huang

Responsive image

Auto-TLDR; Real-World Multimodal Foodlog: A Real-World Foodlog Database for Diet Assistant

Slides Poster Similar

With the increasing health concerns on diet, it's worthwhile to develop an intelligent assistant that can help users eat healthier. Such assistants can automatically give personal advice for the users' diet and generate health reports about eating on a regular basis. To boost the research on such diet assistant, we establish a real-world foodlog database using various methods such as filter, cluster and graph convolutional network. This database is built based on real-world lifelog and medical data, which is named as Real-World Multimodal Foodlog (RWMF). It contains 7500 multimodal pairs, and each pair consists of a food image paired with a line of personal biometrics data (such as Blood Glucose) and a textual food description of food composition paired with a line of food nutrition data. In this paper, we present the detailed procedures for setting up the database. We evaluate the performance of RWMF using different food classification and cross-modal retrieval approaches. We also test the performance of multimodal fusion on RWMF through ablation experiments. The experimental results show that the RWMF database is quite challenging and can be widely used to evaluate the performance of food analysis methods based on multimodal data.

Partially Supervised Multi-Task Network for Single-View Dietary Assessment

Ya Lu, Thomai Stathopoulou, Stavroula Mougiakakou

Responsive image

Auto-TLDR; Food Volume Estimation from a Single Food Image via Geometric Understanding and Semantic Prediction

Slides Poster Similar

Food volume estimation is an essential step in the pipeline of dietary assessment and demands the precise depth estimation of the food surface and table plane. Existing methods based on computer vision require either multi-image input or additional depth maps, reducing convenience of implementation and practical significance. Despite the recent advances in unsupervised depth estimation from a single image, the achieved performance in the case of large texture-less areas needs to be improved. In this paper, we propose a network architecture that jointly performs geometric understanding (i.e., depth prediction and 3D plane estimation) and semantic prediction on a single food image, enabling a robust and accurate food volume estimation regardless of the texture characteristics of the target plane. For the training of the network, only monocular videos with semantic ground truth are required, while the depth map and 3D plane ground truth are no longer needed. Experimental results on two separate food image databases demonstrate that our method performs robustly on texture-less scenarios and is superior to unsupervised networks and structure from motion based approaches, while it achieves comparable performance to fully-supervised methods.

Making Every Label Count: Handling Semantic Imprecision by Integrating Domain Knowledge

Clemens-Alexander Brust, Björn Barz, Joachim Denzler

Responsive image

Auto-TLDR; Class Hierarchies for Imprecise Label Learning and Annotation eXtrapolation

Slides Poster Similar

Noisy data, crawled from the web or supplied by volunteers such as Mechanical Turkers or citizen scientists, is considered an alternative to professionally labeled data. There has been research focused on mitigating the effects of label noise. It is typically modeled as inaccuracy, where the correct label is replaced by an incorrect label from the same set. We consider an additional dimension of label noise: imprecision. For example, a non-breeding snow bunting is labeled as a bird. This label is correct, but not as precise as the task requires. Standard softmax classifiers cannot learn from such a weak label because they consider all classes mutually exclusive, which non-breeding snow bunting and bird are not. We propose CHILLAX (Class Hierarchies for Imprecise Label Learning and Annotation eXtrapolation), a method based on hierarchical classification, to fully utilize labels of any precision. Experiments on noisy variants of NABirds and ILSVRC2012 show that our method outperforms strong baselines by as much as 16.4 percentage points, and the current state of the art by up to 3.9 percentage points.

Uncertainty-Aware Data Augmentation for Food Recognition

Eduardo Aguilar, Bhalaji Nagarajan, Rupali Khatun, Marc Bolaños, Petia Radeva

Responsive image

Auto-TLDR; Data Augmentation for Food Recognition Using Epistemic Uncertainty

Slides Poster Similar

Food recognition has recently attracted attention of many researchers. However, high food ambiguity, inter-class variability and intra-class similarity define a real challenge for the Deep learning and Computer Vision algorithms. In order to improve their performance, it is necessary to better understand what the model learns and, from this, to determine the type of data that should be additionally included for being the most beneficial to the training procedure. In this paper, we propose a new data augmentation strategy that estimates and uses the epistemic uncertainty to guide the model training. The method follows an active learning framework, where the new synthetic images are generated from the hard to classify real ones present in the training data based on the epistemic uncertainty. Hence, it allows the food recognition algorithm to focus on difficult images in order to learn their discriminatives features. On the other hand, avoiding data generation from images that do not contribute to the recognition makes it faster and more efficient. We show that the proposed method allows to improve food recognition and provides a better trade-off between micro- and macro-recall measures.

A Systematic Investigation on End-To-End Deep Recognition of Grocery Products in the Wild

Marco Leo, Pierluigi Carcagni, Cosimo Distante

Responsive image

Auto-TLDR; Automatic Recognition of Products on grocery shelf images using Convolutional Neural Networks

Slides Poster Similar

Automatic recognition of products on grocery shelf images is a new and attractive topic in computer vision and machine learning since, it can be exploited in different application areas. This paper introduces a complete end-to-end pipeline (without preliminary radiometric and spatial transformations usually involved while dealing with the considered issue) and it provides a systematic investigation of recent machine learning models based on convolutional neural networks for addressing the product recognition task by exploiting the proposed pipeline on a recent challenging grocery product dataset. The investigated models were never been used in this context: they derive from the successful and more generic object recognition task and have been properly tuned to address this specific issue. Besides, also ensembles of nets built by most advanced theoretical fundaments have been taken into account. Gathered classification results were very encouraging since the recognition accuracy has been improved up to 15\% with respect to the leading approaches in the state of art on the same dataset. A discussion about the pros and cons of the investigated solutions are discussed by paving the path towards new research lines.

Self-Supervised Learning for Astronomical Image Classification

Ana Martinazzo, Mateus Espadoto, Nina S. T. Hirata

Responsive image

Auto-TLDR; Unlabeled Astronomical Images for Deep Neural Network Pre-training

Slides Poster Similar

In Astronomy, a huge amount of image data is generated daily by photometric surveys, which scan the sky to collect data from stars, galaxies and other celestial objects. In this paper, we propose a technique to leverage unlabeled astronomical images to pre-train deep convolutional neural networks, in order to learn a domain-specific feature extractor which improves the results of machine learning techniques in setups with small amounts of labeled data available. We show that our technique produces results which are in many cases better than using ImageNet pre-training.

Contextual Classification Using Self-Supervised Auxiliary Models for Deep Neural Networks

Sebastian Palacio, Philipp Engler, Jörn Hees, Andreas Dengel

Responsive image

Auto-TLDR; Self-Supervised Autogenous Learning for Deep Neural Networks

Slides Poster Similar

Classification problems solved with deep neural networks (DNNs) typically rely on a closed world paradigm, and optimize over a single objective (e.g., minimization of the cross- entropy loss). This setup dismisses all kinds of supporting signals that can be used to reinforce the existence or absence of particular patterns. The increasing need for models that are interpretable by design makes the inclusion of said contextual signals a crucial necessity. To this end, we introduce the notion of Self-Supervised Autogenous Learning (SSAL). A SSAL objective is realized through one or more additional targets that are derived from the original supervised classification task, following architectural principles found in multi-task learning. SSAL branches impose low-level priors into the optimization process (e.g., grouping). The ability of using SSAL branches during inference, allow models to converge faster, focusing on a richer set of class-relevant features. We equip state-of-the-art DNNs with SSAL objectives and report consistent improvements for all of them on CIFAR100 and Imagenet. We show that SSAL models outperform similar state-of-the-art methods focused on contextual loss functions, auxiliary branches and hierarchical priors.

Multi-Modal Deep Clustering: Unsupervised Partitioning of Images

Guy Shiran, Daphna Weinshall

Responsive image

Auto-TLDR; Multi-Modal Deep Clustering for Unlabeled Images

Slides Poster Similar

The clustering of unlabeled raw images is a daunting task, which has recently been approached with some success by deep learning methods. Here we propose an unsupervised clustering framework, which learns a deep neural network in an end-to-end fashion, providing direct cluster assignments of images without additional processing. Multi-Modal Deep Clustering (MMDC), trains a deep network to align its image embeddings with target points sampled from a Gaussian Mixture Model distribution. The cluster assignments are then determined by mixture component association of image embeddings. Simultaneously, the same deep network is trained to solve an additional self-supervised task. This pushes the network to learn more meaningful image representations and stabilizes the training. Experimental results show that MMDC achieves or exceeds state-of-the-art performance on four challenging benchmarks. On natural image datasets we improve on previous results with significant margins of up to 11% absolute accuracy points, yielding an accuracy of 70% on CIFAR-10 and 61% on STL-10.

A CNN-RNN Framework for Image Annotation from Visual Cues and Social Network Metadata

Tobia Tesan, Pasquale Coscia, Lamberto Ballan

Responsive image

Auto-TLDR; Context-Based Image Annotation with Multiple Semantic Embeddings and Recurrent Neural Networks

Slides Poster Similar

Images represent a commonly used form of visual communication among people. Nevertheless, image classification may be a challenging task when dealing with unclear or non-common images needing more context to be correctly annotated. Metadata accompanying images on social-media represent an ideal source of additional information for retrieving proper neighborhoods easing image annotation task. To this end, we blend visual features extracted from neighbors and their metadata to jointly leverage context and visual cues. Our models use multiple semantic embeddings to achieve the dual objective of being robust to vocabulary changes between train and test sets and decoupling the architecture from the low-level metadata representation. Convolutional and recurrent neural networks (CNNs-RNNs) are jointly adopted to infer similarity among neighbors and query images. We perform comprehensive experiments on the NUS-WIDE dataset showing that our models outperform state-of-the-art architectures based on images and metadata, and decrease both sensory and semantic gaps to better annotate images.

Adversarial Encoder-Multi-Task-Decoder for Multi-Stage Processes

Andre Mendes, Julian Togelius, Leandro Dos Santos Coelho

Responsive image

Auto-TLDR; Multi-Task Learning and Semi-Supervised Learning for Multi-Stage Processes

Similar

In multi-stage processes, decisions occur in an ordered sequence of stages. Early stages usually have more observations with general information (easier/cheaper to collect), while later stages have fewer observations but more specific data. This situation can be represented by a dual funnel structure, in which the sample size decreases from one stage to the other while the information increases. Training classifiers in this scenario is challenging since information in the early stages may not contain distinct patterns to learn (underfitting). In contrast, the small sample size in later stages can cause overfitting. We address both cases by introducing a framework that combines adversarial autoencoders (AAE), multi-task learning (MTL), and multi-label semi-supervised learning (MLSSL). We improve the decoder of the AAE with an MTL component so it can jointly reconstruct the original input and use feature nets to predict the features for the next stages. We also introduce a sequence constraint in the output of an MLSSL classifier to guarantee the sequential pattern in the predictions. Using real-world data from different domains (selection process, medical diagnosis), we show that our approach outperforms other state-of-the-art methods.

Deep Gait Relative Attribute Using a Signed Quadratic Contrastive Loss

Yuta Hayashi, Shehata Allam, Yasushi Makihara, Daigo Muramatsu, Yasushi Yagi

Responsive image

Auto-TLDR; Signal-Contrastive Loss for Gait Attributes Estimation

Similar

This paper presents a deep learning-based method to estimate gait attributes (e.g., stately, cool, relax, etc.). Similarly to the existing studies on relative attribute, human perception-based annotations on the gait attributes are given to pairs of gait videos (i.e., the first one is better, tie, and the second one is better), and the relative annotations are utilized to train a ranking model of the gait attribute. More specifically, we design a Siamese (i.e., two-stream) network which takes a pair of gait inputs and output gait attribute score for each. We then introduce a suitable loss function called a signed contrastive loss to train the network parameters with the relative annotation. Unlike the existing loss functions for learning to rank does not inherent a nice property of a quadratic contrastive loss, the proposed signed quadratic contrastive loss function inherents the nice property. The quantitative evaluation results reveal that the proposed method shows better or comparable accuracies of relative attribute prediction against the baseline methods.

Price Suggestion for Online Second-Hand Items

Liang Han, Zhaozheng Yin, Zhurong Xia, Li Guo, Mingqian Tang, Rong Jin

Responsive image

Auto-TLDR; An Intelligent Price Suggestion System for Online Second-hand Items

Slides Poster Similar

This paper describes an intelligent price suggestion system for online second-hand listings. In contrast to conventional pricing strategies which are employed to a large number of identical products, or to non-identical but similar products such as homes on Airbnb, the proposed system provides price suggestions for online second-hand items which are non-identical and fall into numerous different categories. Moreover, simplifying the item listing process for users is taken into consideration when designing the price suggestion system. Specifically, we design a truncate loss to train a vision-based price suggestion module which mainly takes some vision-based features as input to first classify whether an uploaded item image is qualified for price suggestion, and then offer price suggestions for items with qualified images. For the items with unqualified images, we encourage users to input some text descriptions of the items, and with the text descriptions, we design a multimodal item retrieval module to offer price suggestions. Extensive experiments demonstrate the effectiveness of the proposed system.

Attentive Visual Semantic Specialized Network for Video Captioning

Jesus Perez-Martin, Benjamin Bustos, Jorge Pérez

Responsive image

Auto-TLDR; Adaptive Visual Semantic Specialized Network for Video Captioning

Slides Poster Similar

As an essential high-level task of video understanding topic, automatically describing a video with natural language has recently gained attention as a fundamental challenge in computer vision. Previous models for video captioning have several limitations, such as the existence of gaps in current semantic representations and the inexpressibility of the generated captions. To deal with these limitations, in this paper, we present a new architecture that we callAttentive Visual Semantic Specialized Network(AVSSN), which is an encoder-decoder model based on our Adaptive Attention Gate and Specialized LSTM layers. This architecture can selectively decide when to use visual or semantic information into the text generation process. The adaptive gate makes the decoder to automatically select the relevant information for providing a better temporal state representation than the existing decoders. Besides, the model is capable of learning to improve the expressiveness of generated captions attending to their length, using a sentence-length-related loss function. We evaluate the effectiveness of the proposed approach on the Microsoft Video Description(MSVD) and the Microsoft Research Video-to-Text (MSR-VTT) datasets, achieving state-of-the-art performance with several popular evaluation metrics: BLEU-4, METEOR, CIDEr, and ROUGE_L.

Bridging the Gap between Natural and Medical Images through Deep Colorization

Lia Morra, Luca Piano, Fabrizio Lamberti, Tatiana Tommasi

Responsive image

Auto-TLDR; Transfer Learning for Diagnosis on X-ray Images Using Color Adaptation

Slides Poster Similar

Deep learning has thrived by training on large-scale datasets. However, in many applications, as for medical image diagnosis, getting massive amount of data is still prohibitive due to privacy, lack of acquisition homogeneity and annotation cost. In this scenario transfer learning from natural image collections is a standard practice that attempts to tackle shape, texture and color discrepancy all at once through pretrained model fine-tuning. In this work we propose to disentangle those challenges and design a dedicated network module that focuses on color adaptation. We combine learning from scratch of the color module with transfer learning of different classification backbones obtaining an end-to-end, easy-to-train architecture for diagnostic image recognition on X-ray images. Extensive experiments show how our approach is particularly efficient in case of data scarcity and provides a new path for further transferring the learned color information across multiple medical datasets.

Text Synopsis Generation for Egocentric Videos

Aidean Sharghi, Niels Lobo, Mubarak Shah

Responsive image

Auto-TLDR; Egocentric Video Summarization Using Multi-task Learning for End-to-End Learning

Slides Similar

Mass utilization of body-worn cameras has led to a huge corpus of available egocentric video. Existing video summarization algorithms can accelerate browsing such videos by selecting (visually) interesting shots from them. Nonetheless, since the system user still has to watch the summary videos, browsing large video databases remain a challenge. Hence, in this work, we propose to generate a textual synopsis, consisting of a few sentences describing the most important events in a long egocentric videos. Users can read the short text to gain insight about the video, and more importantly, efficiently search through the content of a large video database using text queries. Since egocentric videos are long and contain many activities and events, using video-to-text algorithms results in thousands of descriptions, many of which are incorrect. Therefore, we propose a multi-task learning scheme to simultaneously generate descriptions for video segments and summarize the resulting descriptions in an end-to-end fashion. We Input a set of video shots and the network generates a text description for each shot. Next, visual-language content matching unit that is trained with a weakly supervised objective, identifies the correct descriptions. Finally, the last component of our network, called purport network, evaluates the descriptions all together to select the ones containing crucial information. Out of thousands of descriptions generated for the video, a few informative sentences are returned to the user. We validate our framework on the challenging UT Egocentric video dataset, where each video is between 3 to 5 hours long, associated with over 3000 textual descriptions on average. The generated textual summaries, including only 5 percent (or less) of the generated descriptions, are compared to groundtruth summaries in text domain using well-established metrics in natural language processing.

Uncertainty-Sensitive Activity Recognition: A Reliability Benchmark and the CARING Models

Alina Roitberg, Monica Haurilet, Manuel Martinez, Rainer Stiefelhagen

Responsive image

Auto-TLDR; CARING: Calibrated Action Recognition with Input Guidance

Slides Similar

Beyond assigning the correct class, an activity recognition model should also to be able to determine, how certain it is in its predictions. We present the first study of how well the confidence values of modern action recognition architectures indeed reflect the probability of the correct outcome and propose a learning-based approach for improving it. First, we extend two popular action recognition datasets with a reliability benchmark in form of the expected calibration error and reliability diagrams. Since our evaluation highlights that confidence values of standard action recognition architectures do not represent the uncertainty well, we introduce a new approach which learns to transform the model output into realistic confidence estimates through an additional calibration network. The main idea of our Calibrated Action Recognition with Input Guidance (CARING) model is to learn an optimal scaling parameter depending on the video representation. We compare our model with the native action recognition networks and the temperature scaling approach - a wide spread calibration method utilized in image classification. While temperature scaling alone drastically improves the reliability of the confidence values, our CARING method consistently leads to the best uncertainty estimates in all benchmark settings.

Towards Tackling Multi-Label Imbalances in Remote Sensing Imagery

Dominik Koßmann, Thorsten Wilhelm, Gernot Fink

Responsive image

Auto-TLDR; Class imbalance in land cover datasets using attribute encoding schemes

Slides Poster Similar

Recent advances in automated image analysis have lead to an increased number of proposed datasets in remote sensing applications. This permits the successful employment of data hungry state-of-the-art deep neural networks. However, the Earth is not covered equally by semantically meaningful classes. Thus, many land cover datasets suffer from a severe class imbalance. We show that by taking appropriate measures, the performance in the minority classes can be improved by up to 30 percent without affecting the performance in the majority classes strongly. Additionally, we investigate the use of an attribute encoding scheme to represent the inherent class hierarchies commonly observed in land cover analysis.

Emerging Relation Network and Task Embedding for Multi-Task Regression Problems

Schreiber Jens, Bernhard Sick

Responsive image

Auto-TLDR; A Comparative Study of Multi-Task Learning for Non-linear Time Series Problems

Slides Poster Similar

Multi-Task learning (MTL) provides state-of-the-art results in many applications of computer vision and natural language processing. In contrast to single-task learning (STL), MTL allows for leveraging knowledge between related tasks improving prediction results on all tasks. However, there is a limited number of comparative studies applied to MTL architectures for regression and time series problems taking recent advances of MTL into account. An intriguing, non-linear time-series problem are day ahead forecasts of the expected power generation for renewable power plants. Therefore, the main contribution of this article is a comparative study of the following recent and relevant MTL architectures: Hard-parameter sharing, cross-stitch network, and sluice network (SN). They are compared to a multi-layer peceptron (MLP) model of similar size in an STL setting. As a additional contribution, we provide a simple, yet practical approach to model task specific information through an embedding layer in an MLP, referred to as task embedding. Further, we contribute a new MTL architecture named emerging relation network (ERN), which can be considered as an extension of the SN. For a solar power dataset, the task embedding achieves the best mean improvement with 8.2%. For two wind and one additional solar dataset, the ERN is the best MTL architecture with improvements up to 11.3%.

Weight Estimation from an RGB-D Camera in Top-View Configuration

Marco Mameli, Marina Paolanti, Nicola Conci, Filippo Tessaro, Emanuele Frontoni, Primo Zingaretti

Responsive image

Auto-TLDR; Top-View Weight Estimation using Deep Neural Networks

Slides Poster Similar

The development of so-called soft-biometrics aims at providing information related to the physical and behavioural characteristics of a person. This paper focuses on bodyweight estimation based on the observation from a top-view RGB-D camera. In fact, the capability to estimate the weight of a person can be of help in many different applications, from health-related scenarios to business intelligence and retail analytics. To deal with this issue, a TVWE (Top-View Weight Estimation) framework is proposed with the aim of predicting the weight. The approach relies on the adoption of Deep Neural Networks (DNNs) that have been trained on depth data. Each network has also been modified in its top section to replace classification with prediction inference. The performance of five state-of-art DNNs has been compared, namely VGG16, ResNet, Inception, DenseNet and Efficient-Net. In addition, a convolutional auto-encoder has also been included for completeness. Considering the limited literature in this domain, the TVWE framework has been evaluated on a new publicly available dataset: “VRAI Weight estimation Dataset”, which also collects, for each subject, labels related to weight, gender, and height. The experimental results have demonstrated that the proposed methods are suitable for this task, bringing different and significant insights for the application of the solution in different domains.

Learning Neural Textual Representations for Citation Recommendation

Thanh Binh Kieu, Inigo Jauregi Unanue, Son Bao Pham, Xuan-Hieu Phan, M. Piccardi

Responsive image

Auto-TLDR; Sentence-BERT cascaded with Siamese and triplet networks for citation recommendation

Slides Poster Similar

With the rapid growth of the scientific literature, manually selecting appropriate citations for a paper is becoming increasingly challenging and time-consuming. While several approaches for automated citation recommendation have been proposed in the recent years, effective document representations for citation recommendation are still elusive to a large extent. For this reason, in this paper we propose a novel approach to citation recommendation which leverages a deep sequential representation of the documents (Sentence-BERT) cascaded with Siamese and triplet networks in a submodular scoring function. To the best of our knowledge, this is the first approach to combine deep representations and submodular selection for a task of citation recommendation. Experiments have been carried out using a popular benchmark dataset -- the ACL Anthology Network corpus -- and evaluated against baselines and a state-of-the-art approach using metrics such as the MRR and F1@k score. The results show that the proposed approach has been able to outperform all the compared approaches in every measured metric.

Exploiting the Logits: Joint Sign Language Recognition and Spell-Correction

Christina Runkel, Stefan Dorenkamp, Hartmut Bauermeister, Michael Möller

Responsive image

Auto-TLDR; A Convolutional Neural Network for Spell-correction in Sign Language Videos

Slides Poster Similar

Machine learning techniques have excelled in the automatic semantic analysis of images, reaching human-level performances on challenging bechmarks. Yet, the semantic analysis of videos remains challenging due to the significantly higher dimensionality of the input data, respectively, the significantly higher need for annotated training examples. By studying the automatic recognition of German sign language videos, we demonstrate that on the relatively scarce training data of 2.800 videos, modern deep learning architectures for video analysis (such as ResNeXt) along with transfer learning on large gesture recognition tasks, can achieve about 75% character accuracy. Considering that this leaves us with a probability of under 25% that a five letter word is spelled correctly, spell-correction systems are crucial for producing readable outputs. The contribution of this paper is to propose a convolutional neural network for spell-correction that expects the softmax outputs of the character recognition network (instead of a misspelled word) as an input. We demonstrate that purely learning on softmax inputs in combination with scarce training data yields overfitting as the network learns the inputs by heart. In contrast, training the network on several variants of the logits of the classification output i.e. scaling by a constant factor, adding of random noise, mixing of softmax and hardmax inputs or purely training on hardmax inputs, leads to better generalization while benefitting from the significant information hidden in these outputs (that have 98% top-5 accuracy), yielding a readable text despite the comparably low character accuracy.

DR2S: Deep Regression with Region Selection for Camera Quality Evaluation

Marcelin Tworski, Stéphane Lathuiliere, Salim Belkarfa, Attilio Fiandrotti, Marco Cagnazzo

Responsive image

Auto-TLDR; Texture Quality Estimation Using Deep Learning

Slides Poster Similar

In this work, we tackle the problem of estimating a camera capability to preserve fine texture details at a given lighting condition. Importantly, our texture preservation measurement should coincide with human perception. Consequently, we formulate our problem as a regression one and we introduce a deep convolutional network to estimate texture quality score. At training time, we use ground-truth quality scores provided by expert human annotators in order to obtain a subjective quality measure. In addition, we propose a region selection method to identify the image regions that are better suited at measuring perceptual quality. Finally, our experimental evaluation shows that our learning-based approach outperforms existing methods and that our region selection algorithm consistently improves the quality estimation.

FC-DCNN: A Densely Connected Neural Network for Stereo Estimation

Dominik Hirner, Friedrich Fraundorfer

Responsive image

Auto-TLDR; FC-DCNN: A Lightweight Network for Stereo Estimation

Slides Poster Similar

We propose a novel lightweight network for stereo estimation. Our network consists of a fully-convolutional densely connected neural network (FC-DCNN) that computes matching costs between rectified image pairs. Our FC-DCNN method learns expressive features and performs some simple but effective post-processing steps. The densely connected layer structure connects the output of each layer to the input of each subsequent layer. This network structure in addition to getting rid of any fully-connected layers leads to a very lightweight network. The output of this network is used in order to calculate matching costs and create a cost-volume. Instead of using time and memory-inefficient cost-aggregation methods such as semi-global matching or conditional random fields in order to improve the result, we rely on filtering techniques, namely median filter and guided filter. By computing a left-right consistency check we get rid of inconsistent values. Afterwards we use a watershed foreground-background segmentation on the disparity image with removed inconsistencies. This mask is then used to refine the final prediction. We show that our method works well for both challenging indoor and outdoor scenes by evaluating it on the Middlebury, KITTI and ETH3D benchmarks respectively.

Large-Scale Historical Watermark Recognition: Dataset and a New Consistency-Based Approach

Xi Shen, Ilaria Pastrolin, Oumayma Bounou, Spyros Gidaris, Marc Smith, Olivier Poncet, Mathieu Aubry

Responsive image

Auto-TLDR; Historical Watermark Recognition with Fine-Grained Cross-Domain One-Shot Instance Recognition

Slides Poster Similar

Historical watermark recognition is a highly practical, yet unsolved challenge for archivists and historians. With a large number of well-defined classes, cluttered and noisy samples, different types of representations, both subtle differences between classes and high intra-class variation, historical watermarks are also challenging for pattern recognition. In this paper, overcoming the difficulty of data collection, we present a large public dataset with more than 6k new photographs, allowing for the first time to tackle at scale the scenarios of practical interest for scholars: one-shot instance recognition and cross-domain one-shot instance recognition amongst more than 16k fine-grained classes. We demonstrate that this new dataset is large enough to train modern deep learning approaches, and show that standard methods can be improved considerably by using mid-level deep features. More precisely, we design both a matching score and a feature fine-tuning strategy based on filtering local matches using spatial consistency. This consistency-based approach provides important performance boost compared to strong baselines. Our model achieves 55\% as top-1 accuracy on our very challenging 16,753-class one-shot cross-domain recognition task, each class described by a single drawing from the classic Briquet catalog. In addition to watermark classification, we show our approach provides promising results on fine-grained sketch-based image retrieval.

Iterative Label Improvement: Robust Training by Confidence Based Filtering and Dataset Partitioning

Christian Haase-Schütz, Rainer Stal, Heinz Hertlein, Bernhard Sick

Responsive image

Auto-TLDR; Meta Training and Labelling for Unlabelled Data

Slides Poster Similar

State-of-the-art, high capacity deep neural networks not only require large amounts of labelled training data, they are also highly susceptible to labelling errors in this data, typically resulting in large efforts and costs and therefore limiting the applicability of deep learning. To alleviate this issue, we propose a novel meta training and labelling scheme that is able to use inexpensive unlabelled data by taking advantage of the generalization power of deep neural networks. We show experimentally that by solely relying on one network architecture and our proposed scheme of combining self-training with pseudolabels, both label quality and resulting model accuracy, can be improved significantly. Our method achieves state-of-the-art results, while being architecture agnostic and therefore broadly applicable. Compared to other methods dealing with erroneous labels, our approach does neither require another network to be trained, nor does it necessarily need an additional, highly accurate reference label set. Instead of removing samples from a labelled set, our technique uses additional sensor data without the need for manual labelling. Furthermore, our approach can be used for semi-supervised learning.

Predicting Chemical Properties Using Self-Attention Multi-Task Learning Based on SMILES Representation

Sangrak Lim, Yong Oh Lee

Responsive image

Auto-TLDR; Self-attention based Transformer-Variant Model for Chemical Compound Properties Prediction

Slides Poster Similar

In the computational prediction of chemical compound properties, molecular descriptors and fingerprints encoded to low dimensional vectors are used. The selection of proper molecular descriptors and fingerprints is both important and challenging as the performance of such models is highly dependent on descriptors. To overcome this challenge, natural language processing models that utilize simplified molecular input line entry system as input were studied, and several transformer variant models achieved superior results when compared with conventional methods. In this study, we explored the structural differences of the transformer-variant model and proposed a new self-attention based model. The representation learning performance of the self-attention module was evaluated in a multi-task learning environment using imbalanced chemical datasets. The experiment results showed that our model achieved competitive outcomes on several benchmark datasets. The source code of our experiment is available at https://github.com/arwhirang/sa-mtl and the dataset is available from the same URL.

ESResNet: Environmental Sound Classification Based on Visual Domain Models

Andrey Guzhov, Federico Raue, Jörn Hees, Andreas Dengel

Responsive image

Auto-TLDR; Environmental Sound Classification with Short-Time Fourier Transform Spectrograms

Slides Poster Similar

Environmental Sound Classification (ESC) is an active research area in the audio domain and has seen a lot of progress in the past years. However, many of the existing approaches achieve high accuracy by relying on domain-specific features and architectures, making it harder to benefit from advances in other fields (e.g., the image domain). Additionally, some of the past successes have been attributed to a discrepancy of how results are evaluated (i.e., on unofficial splits of the UrbanSound8K (US8K) dataset), distorting the overall progression of the field. The contribution of this paper is twofold. First, we present a model that is inherently compatible with mono and stereo sound inputs. Our model is based on simple log-power Short-Time Fourier Transform (STFT) spectrograms and combines them with several well-known approaches from the image domain (i.e., ResNet, Siamese-like networks and attention). We investigate the influence of cross-domain pre-training, architectural changes, and evaluate our model on standard datasets. We find that our model out-performs all previously known approaches in a fair comparison by achieving accuracies of 97.0 % (ESC-10), 91.5 % (ESC-50) and 84.2 % / 85.4 % (US8K mono / stereo). Second, we provide a comprehensive overview of the actual state of the field, by differentiating several previously reported results on the US8K dataset between official or unofficial splits. For better reproducibility, our code (including any re-implementations) is made available.

Smart Inference for Multidigit Convolutional Neural Network Based Barcode Decoding

Duy-Thao Do, Tolcha Yalew, Tae Joon Jun, Daeyoung Kim

Responsive image

Auto-TLDR; Smart Inference for Barcode Decoding using Deep Convolutional Neural Network

Slides Poster Similar

Barcodes are ubiquitous and have been used in most of critical daily activities for decades. However, most of traditional decoders require well-founded barcode under a relatively standard condition. While wilder conditioned barcodes such as underexposed, occluded, blurry, wrinkled and rotated are commonly captured in reality, those traditional decoders show weakness of recognizing. Several works attempted to solve those challenging barcodes, but many limitations still exist. This work aims to solve the decoding problem using deep convolutional neural network with the possibility of running on portable devices. Firstly, we proposed a special modification of inference based on the feature of having checksum and test-time augmentation, named as Smart Inference (SI) in prediction phase of a trained model. SI considerably boosts accuracy and reduces the false prediction for trained models. Secondly, we have created a large practical evaluation dataset of real captured 1D barcode under various challenging conditions to test our methods vigorously, which is publicly available for other researchers. The experiments' results demonstrated the SI effectiveness with the highest accuracy of 95.85% which outperformed many existing decoders on the evaluation set. Finally, we successfully minimized the best model by knowledge distillation to a shallow model which is shown to have high accuracy (90.85%) with good inference speed of 34.2 ms per image on a real edge device.

A Systematic Investigation on Deep Architectures for Automatic Skin Lesions Classification

Pierluigi Carcagni, Marco Leo, Andrea Cuna, Giuseppe Celeste, Cosimo Distante

Responsive image

Auto-TLDR; RegNet: Deep Investigation of Convolutional Neural Networks for Automatic Classification of Skin Lesions

Slides Poster Similar

Computer vision-based techniques are more and more employed in healthcare and medical fields nowadays in order, principally, to be as a support to the experienced medical staff to help them to make a quick and correct diagnosis. One of the hot topics in this arena concerns the automatic classification of skin lesions. Several promising works exist about it, mainly leveraging Convolutional Neural Networks (CNN), but proposed pipeline mainly rely on complex data preprocessing and there is no systematic investigation about how available deep models can actually reach the accuracy needed for real applications. In order to overcome these drawbacks, in this work, an end-to-end pipeline is introduced and some of the most recent Convolutional Neural Networks (CNNs) architectures are included in it and compared on the largest common benchmark dataset recently introduced. To this aim, for the first time in this application context, a new network design paradigm, namely RegNet, has been exploited to get the best models among a population of configurations. The paper introduces a threefold level of contribution and novelty with respect the previous literature: the deep investigation of several CNN architectures driving to a consistent improvement of the lesions recognition accuracy, the exploitation of a new network design paradigm able to study the behavior of populations of models and a deep discussion about pro and cons of each analyzed method paving the path towards new research lines.

Webly Supervised Image-Text Embedding with Noisy Tag Refinement

Niluthpol Mithun, Ravdeep Pasricha, Evangelos Papalexakis, Amit Roy-Chowdhury

Responsive image

Auto-TLDR; Robust Joint Embedding for Image-Text Retrieval Using Web Images

Slides Similar

In this paper, we address the problem of utilizing web images in training robust joint embedding models for the image-text retrieval task. Prior webly supervised approaches directly leverage weakly annotated web images in the joint embedding learning framework. The objective of these approaches would suffer significantly when the ratio of noisy and missing tags associated with the web images is very high. In this regard, we propose a CP decomposition based tensor completion framework to refine the tags of web images by modeling observed ternary inter-relations between the sets of labeled images, tags, and web images as a tensor. To effectively deal with the high ratio of missing entries likely in our case, we incorporate intra-modal correlation as side information in the proposed framework. Our tag refinement approach combined with existing web supervised image-text embedding approaches provide a more principled way for learning the joint embedding models in the presence of significant noise from web data and limited clean labeled data. Experiments on benchmark datasets demonstrate that the proposed approach helps to achieve a significant performance gain in image-text retrieval.

Improving Model Accuracy for Imbalanced Image Classification Tasks by Adding a Final Batch Normalization Layer: An Empirical Study

Veysel Kocaman, Ofer M. Shir, Thomas Baeck

Responsive image

Auto-TLDR; Exploiting Batch Normalization before the Output Layer in Deep Learning for Minority Class Detection in Imbalanced Data Sets

Slides Poster Similar

Some real-world domains, such as Agriculture and Healthcare, comprise early-stage disease indications whose recording constitutes a rare event, and yet, whose precise detection at that stage is critical. In this type of highly imbalanced classification problems, which encompass complex features, deep learning (DL) is much needed because of its strong detection capabilities. At the same time, DL is observed in practice to favor majority over minority classes and consequently suffer from inaccurate detection of the targeted early-stage indications. To simulate such scenarios, we artificially generate skewness (99% vs. 1%) for certain plant types out of the PlantVillage dataset as a basis for classification of scarce visual cues through transfer learning. By randomly and unevenly picking healthy and unhealthy samples from certain plant types to form a training set, we consider a base experiment as fine-tuning ResNet34 and VGG19 architectures and then testing the model performance on a balanced dataset of healthy and unhealthy images. We empirically observe that the initial F1 test score jumps from 0.29 to 0.95 for the minority class upon adding a final Batch Normalization (BN) layer just before the output layer in VGG19. We demonstrate that utilizing an additional BN layer before the output layer in modern CNN architectures has a considerable impact in terms of minimizing the training time and testing error for minority classes in highly imbalanced data sets. Moreover, when the final BN is employed, trying to minimize validation and training losses may not be an optimal way for getting a high F1 test score for minority classes in anomaly detection problems. That is, the network might perform better even if it is not ‘confident’ enough while making a prediction; leading to another discussion about why softmax output is not a good uncertainty measure for DL models.

Enriching Video Captions with Contextual Text

Philipp Rimle, Pelin Dogan, Markus Gross

Responsive image

Auto-TLDR; Contextualized Video Captioning Using Contextual Text

Slides Poster Similar

Understanding video content and generating caption with context is an important and challenging task. Unlike prior methods that typically attempt to generate generic video captions without context, our architecture contextualizes captioning by infusing extracted information from relevant text data. We propose an end-to-end sequence-to-sequence model which generates video captions based on visual input, and mines relevant knowledge such as names and locations from contextual text. In contrast to previous approaches, we do not preprocess the text further, and let the model directly learn to attend over it. Guided by the visual input, the model is able to copy words from the contextual text via a pointer-generator network, allowing to produce more specific video captions. We show competitive performance on the News Video Dataset and, through ablation studies, validate the efficacy of contextual video captioning as well as individual design choices in our model architecture.

Video Face Manipulation Detection through Ensemble of CNNs

Nicolo Bonettini, Edoardo Daniele Cannas, Sara Mandelli, Luca Bondi, Paolo Bestagini, Stefano Tubaro

Responsive image

Auto-TLDR; Face Manipulation Detection in Video Sequences Using Convolutional Neural Networks

Slides Similar

In the last few years, several techniques for facial manipulation in videos have been successfully developed and made available to the masses (i.e., FaceSwap, deepfake, etc.). These methods enable anyone to easily edit faces in video sequences with incredibly realistic results and a very little effort. Despite the usefulness of these tools in many fields, if used maliciously, they can have a significantly bad impact on society (e.g., fake news spreading, cyber bullying through fake revenge porn). The ability of objectively detecting whether a face has been manipulated in a video sequence is then a task of utmost importance. In this paper, we tackle the problem of face manipulation detection in video sequences targeting modern facial manipulation techniques. In particular, we study the ensembling of different trained Convolutional Neural Network (CNN) models. In the proposed solution, different models are obtained starting from a base network (i.e., EfficientNetB4) making use of two different concepts: (i) attention layers; (ii) siamese training. We show that combining these networks leads to promising face manipulation detection results on two publicly available datasets with more than 119000 videos.

Detective: An Attentive Recurrent Model for Sparse Object Detection

Amine Kechaou, Manuel Martinez, Monica Haurilet, Rainer Stiefelhagen

Responsive image

Auto-TLDR; Detective: An attentive object detector that identifies objects in images in a sequential manner

Slides Poster Similar

In this work, we present Detective – an attentive object detector that identifies objects in images in a sequential manner. Our network is based on an encoder-decoder architecture, where the encoder is a convolutional neural network, and the decoder is a convolutional recurrent neural network coupled with an attention mechanism. At each iteration, our decoder focuses on the relevant parts of the image using an attention mechanism, and then estimates the object’s class and the bounding box coordinates. Current object detection models generate dense predictions and rely on post-processing to remove duplicate predictions. Detective is a sparse object detector that generates a single bounding box per object instance. However, training a sparse object detector is challenging, as it requires the model to reason at the instance level and not just at the class and spatial levels. We propose a training mechanism based on the Hungarian Algorithm and a loss that balances the localization and classification tasks. This allows Detective to achieve promising results on the PASCAL VOC object detection dataset. Our experiments demonstrate that sparse object detection is possible and has a great potential for future developments in applications where the order of the objects to be predicted is of interest.

Ballroom Dance Recognition from Audio Recordings

Tomas Pavlin, Jan Cech, Jiri Matas

Responsive image

Auto-TLDR; A CNN-based approach to classify ballroom dances given audio recordings

Slides Poster Similar

We propose a CNN-based approach to classify ten genres of ballroom dances given audio recordings, five latin and five standard, namely Cha Cha Cha, Jive, Paso Doble, Rumba, Samba, Quickstep, Slow Foxtrot, Slow Waltz, Tango and Viennese Waltz. We utilize a spectrogram of an audio signal and we treat it as an image that is an input of the CNN. The classification is performed independently by 5-seconds spectrogram segments in sliding window fashion and the results are then aggregated. The method was tested on following datasets: Publicly available Extended Ballroom dataset collected by Marchand and Peeters, 2016 and two YouTube datasets collected by us, one in studio quality and the other, more challenging, recorded on mobile phones. The method achieved accuracy 93.9%, 96.7% and 89.8% respectively. The method runs in real-time. We implemented a web application to demonstrate the proposed method.

An Evaluation of DNN Architectures for Page Segmentation of Historical Newspapers

Manuel Burghardt, Bernhard Liebl

Responsive image

Auto-TLDR; Evaluation of Backbone Architectures for Optical Character Segmentation of Historical Documents

Slides Poster Similar

One important and particularly challenging step in the optical character recognition of historical documents with complex layouts, such as newspapers, is the separation of text from non-text content (e.g. page borders or illustrations). This step is commonly referred to as page segmentation. While various rule-based algorithms have been proposed, the applicability of Deep Neural Networks for this task recently has gained a lot of attention. In this paper, we perform a systematic evaluation of 11 different published backbone architectures and 9 different tiling and scaling configurations for separating text, tables or table column lines. We also show the influence of the number of labels and the number of training pages on the segmentation quality, which we measure using the Matthews Correlation Coefficient. Our results show that (depending on the task) Inception-ResNet-v2 and EfficientNet backbones work best, vertical tiling is generally preferable to other tiling approaches, and training data that comprises 30 to 40 pages will be sufficient most of the time.

End-To-End Hierarchical Relation Extraction for Generic Form Understanding

Tuan Anh Nguyen Dang, Duc-Thanh Hoang, Quang Bach Tran, Chih-Wei Pan, Thanh-Dat Nguyen

Responsive image

Auto-TLDR; Joint Entity Labeling and Link Prediction for Form Understanding in Noisy Scanned Documents

Slides Poster Similar

Form understanding is a challenging problem which aims to recognize semantic entities from the input document and their hierarchical relations. Previous approaches face a significant difficulty dealing with the complexity of the task, thus treat these objectives separately. To this end, we present a novel deep neural network to jointly perform both Entity Labeling and link prediction in an end-to-end fashion. Our model extends the Multi-stage Attentional U-Net architecture with the Part-Intensity Fields and Part-Association Fields for link prediction, enriching the spatial information flow with the additional supervision from Entity Linking. We demonstrate the effectiveness of the model on the \textit{Form Understanding in Noisy Scanned Documents} \textit{(FUNSD)} dataset, where our method substantially outperforms the original model and state-of-the-art baselines in both Entity Labeling and Entity Linking task.

A Novel Attention-Based Aggregation Function to Combine Vision and Language

Matteo Stefanini, Marcella Cornia, Lorenzo Baraldi, Rita Cucchiara

Responsive image

Auto-TLDR; Fully-Attentive Reduction for Vision and Language

Slides Poster Similar

The joint understanding of vision and language has been recently gaining a lot of attention in both the Computer Vision and Natural Language Processing communities, with the emergence of tasks such as image captioning, image-text matching, and visual question answering. As both images and text can be encoded as sets or sequences of elements - like regions and words - proper reduction functions are needed to transform a set of encoded elements into a single response, like a classification or similarity score. In this paper, we propose a novel fully-attentive reduction method for vision and language. Specifically, our approach computes a set of scores for each element of each modality employing a novel variant of cross-attention, and performs a learnable and cross-modal reduction, which can be used for both classification and ranking. We test our approach on image-text matching and visual question answering, building fair comparisons with other reduction choices, on both COCO and VQA 2.0 datasets. Experimentally, we demonstrate that our approach leads to a performance increase on both tasks. Further, we conduct ablation studies to validate the role of each component of the approach.

Learning from Web Data: Improving Crowd Counting Via Semi-Supervised Learning

Tao Peng, Pengfei Zhu

Responsive image

Auto-TLDR; Semi-supervised Crowd Counting Baseline for Deep Neural Networks

Slides Poster Similar

Deep neural networks need large-scale dataset for better training and evaluation. However collecting and annotating large-scale crowd counting dataset is expensive and challenging. In this work, we exploit unlabeled web crowd image and propose an multi-task framework for boosting crowd counting baseline method through semi-supervision.Based on the observation that the rotation and splitting operations will not change the image crowd counting number,we designed three auxiliary tasks to improve the quality of feature extractors and our framework can be easily extended to other crowd counting baselines. Experiments shows that our semi-supervised learning framework outperforms previous baselines on UCF-QNRF dataset and ShanghaiTech dataset.

Cross-Lingual Text Image Recognition Via Multi-Task Sequence to Sequence Learning

Zhuo Chen, Fei Yin, Xu-Yao Zhang, Qing Yang, Cheng-Lin Liu

Responsive image

Auto-TLDR; Cross-Lingual Text Image Recognition with Multi-task Learning

Slides Poster Similar

This paper considers recognizing texts shown in a source language and translating into a target language, without generating the intermediate source language text image recognition results. We call this problem Cross-Lingual Text Image Recognition (CLTIR). To solve this problem, we propose a multi-task system containing a main task of CLTIR and an auxiliary task of Mono-Lingual Text Image Recognition (MLTIR) simultaneously. Two different sequence to sequence learning methods, a convolution based attention model and a BLSTM model with CTC, are adopted for these tasks respectively. We evaluate the system on a newly collected Chinese-English bilingual movie subtitle image dataset. Experimental results demonstrate the multi-task learning framework performs superiorly in both languages.

Confidence Calibration for Deep Renal Biopsy Immunofluorescence Image Classification

Federico Pollastri, Juan Maroñas, Federico Bolelli, Giulia Ligabue, Roberto Paredes, Riccardo Magistroni, Costantino Grana

Responsive image

Auto-TLDR; A Probabilistic Convolutional Neural Network for Immunofluorescence Classification in Renal Biopsy

Slides Poster Similar

With this work we tackle immunofluorescence classification in renal biopsy, employing state-of-the-art Convolutional Neural Networks. In this setting, the aim of the probabilistic model is to assist an expert practitioner towards identifying the location pattern of antibody deposits within a glomerulus. Since modern neural networks often provide overconfident outputs, we stress the importance of having a reliable prediction, demonstrating that Temperature Scaling, a recently introduced re-calibration technique, can be successfully applied to immunofluorescence classification in renal biopsy. Experimental results demonstrate that the designed model yields good accuracy on the specific task, and that Temperature Scaling is able to provide reliable probabilities, which are highly valuable for such a task given the low inter-rater agreement.

Conditional Multi-Task Learning for Plant Disease Identification

Sue Han Lee, Herve Goëau, Pierre Bonnet, Alexis Joly

Responsive image

Auto-TLDR; A conditional multi-task learning approach for plant disease identification

Slides Poster Similar

Several recent studies have proposed an automatic plant disease identification system based on deep learning. Although successful, these approaches are generally based on learned classification models with target classes of joint host species-disease pairs that may not allow optimal use of the available information. This is due to the fact that they require distinguishing between similar host species or diseases. In fact, these approaches have limited scalability because the size of a network gradually increases as new classes are added, even if information on host species or diseases is already available. This constraint is all the more important as it can be difficult to collect/establish a specific list of all diseases for each host plant species in an actual application. In this paper, we address the problems by proposing a new conditional multi-task learning (CMTL) approach which allows the distribution of host species and disease characteristics learned simultaneously with a conditional link between them. This conditioning is formed in such a way that the knowledge to infer the prediction of one concept (the diseases) depends on the other concept (the host species), which corresponds to the way plant pathologists used to infer the diseases of the host species. We show that our approach can improve the performance of plant disease identification compared to the usual species-disease pair modeling in the previous studies. Meanwhile, we also compose a new dataset on plant disease identification that could serve as an important benchmark in this field.

How Unique Is a Face: An Investigative Study

Michal Balazia, S L Happy, Francois Bremond, Antitza Dantcheva

Responsive image

Auto-TLDR; Uniqueness of Face Recognition: Exploring the Impact of Factors such as image resolution, feature representation, database size, age and gender

Slides Poster Similar

Face recognition has been widely accepted as a means of identification in applications ranging from border control to security in the banking sector. Surprisingly, while widely accepted, we still lack the understanding of the uniqueness or distinctiveness of face as a biometric characteristic. In this work, we study the impact of factors such as image resolution, feature representation, database size, age and gender on uniqueness denoted by the Kullback-Leibler divergence between genuine and impostor distributions. Towards understanding the impact, we present experimental results on the datasets AT&T, LFW, IMDb-Face, as well as ND-TWINS, with the feature extraction algorithms VGGFace, VGG16, ResNet50, InceptionV3, MobileNet and DenseNet121, that reveal the quantitative impact of the named factors. While these are early results, our findings indicate the need for a better understanding of the concept of biometric uniqueness and its implication on face recognition.

Multimodal Side-Tuning for Document Classification

Stefano Zingaro, Giuseppe Lisanti, Maurizio Gabbrielli

Responsive image

Auto-TLDR; Side-tuning for Multimodal Document Classification

Slides Poster Similar

In this paper, we propose to exploit the side-tuning framework for multimodal document classification. Side-tuning is a methodology for network adaptation recently introduced to solve some of the problems related to previous approaches. Thanks to this technique it is actually possible to overcome model rigidity and catastrophic forgetting of transfer learning by fine-tuning. The proposed solution uses off-the-shelf deep learning architectures leveraging the side-tuning framework to combine a base model with a tandem of two side networks. We show that side-tuning can be successfully employed also when different data sources are considered, e.g. text and images in document classification. The experimental results show that this approach pushes further the limit for document classification accuracy with respect to the state of the art.

The DeepHealth Toolkit: A Unified Framework to Boost Biomedical Applications

Michele Cancilla, Laura Canalini, Federico Bolelli, Stefano Allegretti, Salvador Carrión, Roberto Paredes, Jon Ander Gómez, Simone Leo, Marco Enrico Piras, Luca Pireddu, Asaf Badouh, Santiago Marco-Sola, Lluc Alvarez, Miquel Moreto, Costantino Grana

Responsive image

Auto-TLDR; DeepHealth Toolkit: An Open Source Deep Learning Toolkit for Cloud Computing and HPC

Slides Poster Similar

Given the overwhelming impact of machine learning on the last decade, several libraries and frameworks have been developed in recent years to simplify the design and training of neural networks, providing array-based programming, automatic differentiation and user-friendly access to hardware accelerators. None of those tools, however, was designed with native and transparent support for Cloud Computing or heterogeneous High-Performance Computing (HPC). The DeepHealth Toolkit is an open source deep learning toolkit aimed at boosting productivity of data scientists operating in the medical field by providing a unified framework for the distributed training of neural networks, that is able to leverage hybrid HPC and Cloud environments in a way transparent to the user. The toolkit is composed of a computer vision library, a deep learning library, and a front-end for non-expert users; all of the components are focused on the medical domain, but they are general purpose and can be applied to any other field. In this paper, the principles driving the design of the DeepHealth libraries are described, along with details about the implementation and the interaction between the different elements composing the toolkit. Finally, experiments on common benchmarks prove the efficiency of each separate component, and of the DeepHealth Toolkit overall.

Improving Word Recognition Using Multiple Hypotheses and Deep Embeddings

Siddhant Bansal, Praveen Krishnan, C. V. Jawahar

Responsive image

Auto-TLDR; EmbedNet: fuse recognition-based and recognition-free approaches for word recognition using learning-based methods

Slides Poster Similar

We propose to fuse recognition-based and recognition-free approaches for word recognition using learning-based methods. For this purpose, results obtained using a text recognizer and deep embeddings (generated using an End2End network) are fused. To further improve the embeddings, we propose EmbedNet, it uses triplet loss for training and learns an embedding space where the embedding of the word image lies closer to its corresponding text transcription’s embedding. This updated embedding space helps in choosing the correct prediction with higher confidence. To further improve the accuracy, we propose a plug-and-play module called Confidence based Accuracy Booster (CAB). It takes in the confidence scores obtained from the text recognizer and Euclidean distances between the embeddings and generates an updated distance vector. This vector has lower distance values for the correct words and higher distance values for the incorrect words. We rigorously evaluate our proposed method systematically on a collection of books that are in the Hindi language. Our method achieves an absolute improvement of around 10% in terms of word recognition accuracy.

RMS-Net: Regression and Masking for Soccer Event Spotting

Matteo Tomei, Lorenzo Baraldi, Simone Calderara, Simone Bronzin, Rita Cucchiara

Responsive image

Auto-TLDR; An Action Spotting Network for Soccer Videos

Slides Poster Similar

The recently proposed action spotting task consists in finding the exact timestamp in which an event occurs. This task fits particularly well for soccer videos, where events correspond to salient actions strictly defined by soccer rules (a goal occurs when the ball crosses the goal line). In this paper, we devise a lightweight and modular network for action spotting, which can simultaneously predict the event label and its temporal offset using the same underlying features. We enrich our model with two training strategies: the first one for data balancing and uniform sampling, the second for masking ambiguous frames and keeping the most discriminative visual cues. When tested on the SoccerNet dataset and using standard features, our full proposal exceeds the current state of the art by 3 Average-mAP points. Additionally, it reaches a gain of more than 10 Average-mAP points on the test set when fine-tuned in combination with a strong 2D backbone.