Human observers engage in selective information uptake when classifying visual patterns. The same is true of deep neural networks, which currently constitute the best performing artificial vision systems. Our goal is to examine the congruence, or lack thereof, in the information-gathering strategies of the two systems. We have operationalized our investigation as a character recognition task. We have used eye-tracking to assay the spatial distribution of information hotspots for humans via fixation maps and an activation mapping technique for obtaining analogous distributions for deep networks through visualization maps. Qualitative comparison between visualization maps and fixation maps reveals an interesting correlate of congruence. The deep learning model considered similar regions in character, which humans have fixated in the case of correctly classified characters. On the other hand, when the focused regions are different for humans and deep nets, the characters are typically misclassified by the latter. Hence, we propose to use the visual fixation maps obtained from the eye-tracking experiment as a supervisory input to align the model's focus on relevant character regions. We find that such supervision improves the model's performance significantly and does not require any additional parameters. This approach has the potential to find applications in diverse domains such as medical analysis and surveillance in which explainability helps to determine system fidelity.
It is difficult for people to interpret the decision-making in the inference process of deep neural networks. Visual explanation is one method for interpreting the decision-making of deep learning. It analyzes the decision-making of 2D CNNs by visualizing an attention map that highlights discriminative regions. Visual explanation for interpreting the decision-making process in video recognition is more difficult because it is necessary to consider not only spatial but also temporal information, which is different from the case of still images. In this paper, we propose a visual explanation met…
Visual explanation enables human to understand the decision making of Deep Convolutional Neural Network (CNN), but it is insufficient to contribute the performance improvement. In this paper, we focus on the attention map for visual explanation, which represents high response value as the important region in image recognition. This region significantly improves the performance of CNN by introducing an attention mechanism that focuses on a specific region in an image. In this work, we propose Attention Branch Network (ABN), which extends the top-down visual explanation model by introducing a br…
In this paper, Gated-ViGAT, an efficient approach for video event recognition, utilizing bottom-up (object) information, a new frame sampling policy and a gating mechanism is proposed. Specifically, the frame sampling policy uses weighted in-degrees (WiDs), derived from the adjacency matrices of graph attention networks (GATs), and a dissimilarity measure to select the most salient and at the same time diverse frames representing the event in the video. Additionally, the proposed gating mechanism fetches the selected frames sequentially, and commits early-exiting when an adequately confident d…
Explainable artificial intelligence has been gaining attention in the past few years. However, most existing methods are based on gradients or intermediate features, which are not directly involved in the decision-making process of the classifier. In this paper, we propose a slot attention-based classifier called SCOUTER for transparent yet accurate classification. Two major differences from other attention-based methods include: (a) SCOUTER's explanation is involved in the final confidence for each category, offering more intuitive interpretation, and (b) all the categories have their corresp…
Fraunhofer Institute for Integrated Circuits（弗劳恩霍夫集成电路研究所）；University of Bamberg（班贝格大学）
Neural networks with high performance can still be biased towards non-relevant features. However, reliability and robustness is especially important for high-risk fields such as clinical pain treatment. We therefore propose a verification pipeline, which consists of three steps. First, we classify facial expressions with a neural network. Next, we apply layer-wise relevance propagation to create pixel-based explanations. Finally, we quantify these visual explanations based on a bounding-box method with respect to facial regions. Although our results show that the neural network achieves state-…
Recently, face recognition systems have demonstrated remarkable performances and thus gained a vital role in our daily life. They already surpass human face verification accountability in many scenarios. However, they lack explanations for their predictions. Compared to human operators, typical face recognition network system generate only binary decisions without further explanation and insights into those decisions. This work focuses on explanations for face recognition systems, vital for developers and operators. First, we introduce a confidence score for those systems based on facial featu…
Black-box Attacks on Image Activity Prediction and its Natural Language Explanations
Idiap Research Institute（伊迪亚普研究所）；University of Perugia（佩鲁贾大学）；École Polytechnique Fédérale de Lausanne（洛桑联邦理工学院）；Queen Mary University of London（伦敦玛丽女王大学）
State Key Laboratory for Novel Software Technology（计算机软件新技术国家重点实验室）；Nanjing University（南京大学）
Multi-label image recognition is a challenging computer vision task of practical use. Progresses in this area, however, are often characterized by complicated methods, heavy computations, and lack of intuitive explanations. To effectively capture different spatial regions occupied by objects from different categories, we propose an embarrassingly simple module, named class-specific residual attention (CSRA). CSRA generates class-specific features for every category by proposing a simple spatial attention score, and then combines it with the class-agnostic average pooling feature. CSRA achieves…
University of Munich（慕尼黑大学）；Siemens AG, Corporate Technology（西门子股份公司，企业技术部）
Capsule Networks, as alternatives to Convolutional Neural Networks, have been proposed to recognize objects from images. The current literature demonstrates many advantages of CapsNets over CNNs. However, how to create explanations for individual classifications of CapsNets has not been well explored. The widely used saliency methods are mainly proposed for explaining CNN-based classifications; they create saliency map explanations by combining activation values and the corresponding gradients, e.g., Grad-CAM. These saliency methods require a specific architecture of the underlying classifiers…
Information Technologies Institute/Centre for Research
In this paper a pure-attention bottom-up approach, called ViGAT, that utilizes an object detector together with a Vision Transformer (ViT) backbone network to derive object and frame features, and a head network to process these features for the task of event recognition and explanation in video, is proposed. The ViGAT head consists of graph attention network (GAT) blocks factorized along the spatial and temporal dimensions in order to capture effectively both local and long-term dependencies between objects or frames. Moreover, using the weighted in-degrees (WiDs) derived from the adjacency m…
Explainable Image Recognition via Enhanced Slot-attention Based Classifier
Institute of Datability Science (IDS), Osaka University（大阪大学数据科学研究所）；Premium Research Institute for Human Metaverse Medicine (WPI-PRIMe), Osaka University（大阪大学人类元宇宙医学高级研究院）
Utrecht University（乌得勒支大学）；University of Essex（埃塞克斯大学）；London South Bank University（伦敦南岸大学）
Deep convolutional networks are widely used in video action recognition. 3D convolutions are one prominent approach to deal with the additional time dimension. While 3D convolutions typically lead to higher accuracies, the inner workings of the trained models are more difficult to interpret. We focus on creating human-understandable visual explanations that represent the hierarchical parts of spatio-temporal networks. We introduce Class Feature Pyramids, a method that traverses the entire network structure and incrementally discovers kernels at different network depths that are informative for…
Nagoya University（名古屋大学）；Hitotsubashi University（一桥大学）；Ritsumeikan University（立命馆大学）
Recent studies have achieved outstanding success in explaining 2D image recognition ConvNets. On the other hand, due to the computation cost and complexity of video data, the explanation of 3D video recognition ConvNets is relatively less studied. In this paper, we present a 3D ACE (Automatic Concept-based Explanation) framework for interpreting 3D ConvNets. In our approach: (1) videos are represented using high-level supervoxels, which is straightforward for human to understand; and (2) the interpreting framework estimates a score for each voxel, which reflects its importance in the decision …
Visual explanations are logical arguments based on visual features that justify the predictions made by neural networks. Current modes of visual explanations answer questions of the form $`Why \text{ } P?'$. These $Why$ questions operate under broad contexts thereby providing answers that are irrelevant in some cases. We propose to constrain these $Why$ questions based on some context $Q$ so that our explanations answer contrastive questions of the form $`Why \text{ } P, \text{} rather \text{ } than \text{ } Q?'$. In this paper, we formalize the structure of contrastive visual explanations for…
UCLA（加州大学洛杉矶分校）；University of Michigan（密歇根大学）；Oregon State University（俄勒冈州立大学）；Beijing Institute for General AI (BIGAI)（北京通用人工智能研究院）；Tsinghua University（清华大学）；Peking University（北京大学）
We propose CX-ToM, short for counterfactual explanations with theory-of mind, a new explainable AI (XAI) framework for explaining decisions made by a deep convolutional neural network (CNN). In contrast to the current methods in XAI that generate explanations as a single shot response, we pose explanation as an iterative communication process, i.e. dialog, between the machine and human user. More concretely, our CX-ToM framework generates sequence of explanations in a dialog by mediating the differences between the minds of machine and human user. To do this, we use Theory of Mind (ToM) which …
Preferred Networks；The University of Tokyo（东京大学）；RIKEN（理化学研究所）
This study addresses generating counterfactual explanations with multimodal information. Our goal is not only to classify a video into a specific category, but also to provide explanations on why it is not categorized to a specific class with combinations of visual-linguistic information. Requirements that the expected output should satisfy are referred to as counterfactuality in this paper: (1) Compatibility of visual-linguistic explanations, and (2) Positiveness/negativeness for the specific positive/negative class. Exploiting a spatio-temporal region (tube) and an attribute as visual and li…
3D Convolutional Neural Network (3D CNN) captures spatial and temporal information on 3D data such as video sequences. However, due to the convolution and pooling mechanism, the information loss seems unavoidable. To improve the visual explanations and classification in 3D CNN, we propose two approaches; i) aggregate layer-wise global to local (global-local) discrete gradients using trained 3DResNext network, and ii) implement attention gating network to improve the accuracy of the action recognition. The proposed approach intends to show the usefulness of every layer termed as global-local at…
University of Twente（特文特大学）；University of Duisburg-Essen（杜伊斯堡-埃森大学）
Image recognition with prototypes is considered an interpretable alternative for black box deep learning models. Classification depends on the extent to which a test image "looks like" a prototype. However, perceptual similarity for humans can be different from the similarity learned by the classification model. Hence, only visualising prototypes can be insufficient for a user to understand what a prototype exactly represents, and why the model considers a prototype and an image to be similar. We address this ambiguity and argue that prototypes should be explained. We improve interpretability …
I describe an approach to similarity motivated by Bayesian methods. This yields a similarity function that is learnable using a standard Bayesian methods. The relationship of the approach to variable kernel and variable metric methods is discussed. The approach is related to variable kernel Experimental results on character recognition and 3D object recognition are presented..
CA-Stream: Attention-based pooling for interpretable image recognition
Centrale Marseille（马赛中央理工学院）；Aix Marseille Univ（艾克斯-马赛大学）；CNRS（法国国家科学研究中心）；Institute of Intelligent Software（智能软件研究所）；Institute of Advanced Research on Artificial Intelligence (IARAI)（高级人工智能研究所（IARAI））
Nagoya Institute of Technology（名古屋工业大学）；Chubu University（中部大学）
In this paper we propose an extension of the Attention Branch Network (ABN) by using instance segmentation for generating sharper attention maps for action recognition. Methods for visual explanation such as Grad-CAM usually generate blurry maps which are not intuitive for humans to understand, particularly in recognizing actions of people in videos. Our proposed method, Object-ABN, tackles this issue by introducing a new mask loss that makes the generated attention maps close to the instance segmentation result. Further the PC loss and multiple attention maps are introduced to enhance the sha…
Understanding how cities visually differ from each others is interesting for planners, residents, and historians. We investigate the interpretation of deep features learned by convolutional neural networks (CNNs) for city recognition. Given a trained city recognition network, we first generate weighted masks using the known Grad-CAM technique and to select the most discriminate regions in the image. Since the image classification label is the city name, it contains no information of objects that are class-discriminate, we investigate the interpretability of deep representations with two method…
This paper describes a new feature set for use in the recognition of on-line handwritten Devanagari script based on Fuzzy Directional Features. Experiments are conducted for the automatic recognition of isolated handwritten character primitives (sub-character units). Initially we describe the proposed feature set, called the Fuzzy Directional Features (FDF) and then show how these features can be effectively utilized for writer independent character recognition. Experimental results show that FDF set perform well for writer independent data set at stroke level recognition. The main contributio…
The main challenge in on-line handwritten character recognition in Indian lan- guage is the large size of the character set, larger similarity between different characters in the script and the huge variation in writing style. In this paper we propose a framework for on-line handwitten script recognition taking cues from speech signal processing literature. The framework is based on identify- ing strokes, which in turn lead to recognition of handwritten on-line characters rather that the conventional character identification. Though the framework is described for Devanagari script, the framewo…
This paper describes a geometry based technique for feature extraction applicable to segmentation-based word recognition systems. The proposed system extracts the geometric features of the character contour. This features are based on the basic line types that forms the character skeleton. The system gives a feature vector as its output. The feature vectors so generated from a training set, were then used to train a pattern recognition engine based on Neural Networks so that the system can be benchmarked.
M. S. Ramaiah Institute of Technology（M. S. 拉迈亚理工学院）
Intensive research has been done on optical character recognition ocr and a large number of articles have been published on this topic during the last few decades. Many commercial OCR systems are now available in the market, but most of these systems work for Roman, Chinese, Japanese and Arabic characters. There are no sufficient number of works on Indian language character recognition especially Kannada script among 12 major scripts in India. This paper presents a review of existing work on printed Kannada script and their results. The characteristics of Kannada script and Kannada Character R…
Transliteration involves transformation of one script to another based on phonetic similarities between the characters of two distinctive scripts. In this paper, we present a novel technique for automatic transliteration of Devanagari script using character recognition. One of the first tasks performed to isolate the constituent characters is segmentation. Line segmentation methodology in this manuscript discusses the case of overlapping lines. Character segmentation algorithm is designed to segment conjuncts and separate shadow characters. Presented shadow character segmentation scheme employ…
Handwriting recognition has been one of the most fascinating and challenging research areas in field of image processing and pattern recognition. It contributes enormously to the improvement of automation process. In this paper, a system for recognition of unconstrained handwritten Malayalam characters is proposed. A database of 10,000 character samples of 44 basic Malayalam characters is used in this work. A discriminate feature set of 64 local and 4 global features are used to train and test SVM classifier and achieved 92.24% accuracy
Human eye can see and read what is written or displayed either in natural handwriting or in printed format. The same work in case the machine does is called handwriting recognition. Handwriting recognition can be broken down into two categories: off-line and on-line. ...
In the paper, we combined DSP processor with image processing algorithm and studied the method of water meter character recognition. We collected water meter image through camera at a fixed angle, and the projection method is used to recognize those digital images. The experiment results show that the method can recognize the meter characters accurately and artificial meter reading is replaced by automatic digital recognition, which improves working efficiency.
Recently, increasing attention has been drawn to the internal mechanisms of convolutional neural networks, and the reason why the network makes specific decisions. In this paper, we develop a novel post-hoc visual explanation method called Score-CAM based on class activation mapping. Unlike previous class activation mapping based approaches, Score-CAM gets rid of the dependence on gradients by obtaining the weight of each activation map through its forward passing score on target class, the final result is obtained by a linear combination of weights and activation maps. We demonstrate that Sco…
Rajagiri School of Engineering & Technology（拉贾吉里工程技术学院）
Handwritten character recognition is one of the most challenging and ongoing areas of research in the field of pattern recognition. HCR research is matured for foreign languages like Chinese and Japanese but the problem is much more complex for Indian languages. The problem becomes even more complicated for South Indian languages due to its large character set and the presence of vowels modifiers and compound characters. This paper provides an overview of important contributions and advances in offline as well as online handwritten character recognition of Malayalam scripts.
A visual counterfactual explanation replaces image regions in a query image with regions from a distractor image such that the system's decision on the transformed image changes to the distractor class. In this work, we present a novel framework for computing visual counterfactual explanations based on two key ideas. First, we enforce that the replaced and replacer regions contain the same semantic part, resulting in more semantically consistent explanations. Second, we use multiple distractor images in a computationally efficient way and obtain more discriminative explanations with fewer regi…
Technische Universität Dortmund（多特蒙德工业大学）；University of Liberal Arts Bangladesh（孟加拉自由艺术大学）
Automatic image and digit recognition is a computationally challenging task for image processing and pattern recognition, requiring an adequate appreciation of the syntactic and semantic importance of the image for the identification ofthe handwritten digits. Image and Pattern Recognition has been identified as one of the driving forces in the research areas because of its shifting of different types of applications, such as safety frameworks, clinical frameworks, diversion, and so on.In this study, for recognition, we implemented a hybrid neural network model that is capable of recognizing th…
Character identification plays a vital role in the contemporary world of Image processing. It can solve many composite problems and makes humans work easier. An instance is Handwritten Character detection. Handwritten recognition is not a novel expertise, but it has not gained community notice until Now. The eventual aim of designing Handwritten Character recognition structure with an accurateness rate of 100% is pretty illusionary. Tamil Handwritten Character recognition system uses the Neural Networks to distinguish them. Neural Network and structural characteristics are used to instruct and…
Meghnad Saha Institute of Technology（梅格纳德·萨哈技术学院）；Jadavpur University（贾达普大学）
In this paper a scheme for offline Handwritten Devnagari Character Recognition is proposed, which uses different feature extraction methodologies and recognition algorithms. The proposed system assumes no constraints in writing style or size. First the character is preprocessed and features namely : Chain code histogram and moment invariant features are extracted and fed to Multilayer Perceptrons as a preliminary recognition step. Finally the results of both MLP's are combined using weighted majority scheme. The proposed system is tested on 1500 handwritten devnagari character database collect…
Handwritten character recognition is an active research challenge,especially for Indian scripts. This paper deals with handwritten Malayalam, with a complete set of basic characters, vowel and consonant signs and compound characters that may be present in the script. Spatial domain features suitable for recognition are chosen in this work. For classification, k-NN, SVM and ELM are employed
