In a recent decade, ImageNet has become the most notable and powerful benchmark database in computer vision and machine learning community. As ImageNet has emerged as a representative benchmark for evaluating the performance of novel deep learning models, its evaluation tends to include only quantitative measures such as error rate, rather than qualitative analysis. Thus, there are few studies that analyze the failure cases of deep learning models in ImageNet, though there are numerous works analyzing the networks themselves and visualizing them. In this abstract, we qualitatively analyze the failure cases of ImageNet classification results from recent deep learning model, and categorize these cases according to the certain image patterns. Through this failure analysis, we believe that it can be discovered what the final challenges are in ImageNet database, which the current deep learning model is still vulnerable to.
Convolutional neural networks (ConvNets) are the actual standard for image recognizement and classification. On the present work we develop a Computer Aided-Diagnosis (CAD) system using ConvNets to classify a x-rays chest images dataset in two groups: Normal and Pneumonia. The study uses ConvNets models available on the PyTorch platform: AlexNet, SqueezeNet, ResNet and Inception. We initially use three training styles: complete from scratch using random initialization, using a pre-trained ImageNet model training only the last layer adapted to our problem (transfer learning) and a pre-trained m…
Beijing University of Posts and Telecommunications（北京邮电大学）；ByteDance AI Lab（字节跳动人工智能实验室）
Although convolutional networks (ConvNets) have enjoyed great success in computer vision (CV), it suffers from capturing global information crucial to dense prediction tasks such as object detection and segmentation. In this work, we innovatively propose ConTNet (ConvolutionTransformer Network), combining transformer with ConvNet architectures to provide large receptive fields. Unlike the recently-proposed transformer-based models (e.g., ViT, DeiT) that are sensitive to hyper-parameters and extremely dependent on a pile of data augmentations when trained from scratch on a midsize dataset (e.g.…
Large Convolutional Network models have recently demonstrated impressive classification performance on the ImageNet benchmark. However there is no clear understanding of why they perform so well, or how they might be improved. In this paper we address both issues. We introduce a novel visualization technique that gives insight into the function of intermediate feature layers and the operation of the classifier. We also perform an ablation study to discover the performance contribution from different model layers. This enables us to find model architectures that outperform Krizhevsky \etal on t…
Wrocław University of Science and Technology（弗罗茨瓦夫理工大学）
Convolutional neural networks applied for real-world classification tasks need to recognize inputs that are far or out-of-distribution (OoD) with respect to the known or training data. To achieve this, many methods estimate class-conditional posterior probabilities and use confidence scores obtained from the posterior distributions. Recent works propose to use multivariate Gaussian distributions as models of posterior distributions at different layers of the CNN (i.e., for low- and upper-level features), which leads to the confidence scores based on the Mahalanobis distance. However, this proc…
Why is it that we can recognize object identity and 3D shape from line drawings, even though they do not exist in the natural world? This paper hypothesizes that the human visual system perceives line drawings as if they were approximately realistic images. Moreover, the techniques of line drawing are chosen to accurately convey shape to a human observer. Several implications and variants of this hypothesis are explored.
Tsinghua University（清华大学）；Huawei Singapore Research Center（华为新加坡研究中心）
Recently, it was found that many real-world examples without intentional modifications can fool machine learning models, and such examples are called "natural adversarial examples". ImageNet-A is a famous dataset of natural adversarial examples. By analyzing this dataset, we hypothesized that large, cluttered and/or unusual background is an important reason why the images in this dataset are difficult to be classified. We validated the hypothesis by reducing the background influence in ImageNet-A examples with object detection techniques. Experiments showed that the object detection models wit…
Why does my medical AI look at pictures of birds? Exploring the efficacy of transfer learning across domain boundaries
University Duisburg-Essen（杜伊斯堡-埃森大学）；University Medicine Essen (AöR)（埃森大学医学中心（公法机构））；Ruhr University Bochum（波鸿鲁尔大学）；Monash University（莫纳什大学）；Cancer Research Center Cologne Essen (CCCE)（科隆埃森癌症研究中心）；TU Dortmund University（多特蒙德工业大学）
Google Research（谷歌研究院）；University of Michigan（密歇根大学）
Previous work has proposed many new loss functions and regularizers that improve test accuracy on image classification tasks. However, it is not clear whether these loss functions learn better representations for downstream tasks. This paper studies how the choice of training objective affects the transferability of the hidden representations of convolutional neural networks trained on ImageNet. We show that many objectives lead to statistically significant improvements in ImageNet accuracy over vanilla softmax cross-entropy, but the resulting fixed feature extractors transfer substantially wo…
Is ImageNet worth 1 video? Learning strong image encoders from 1 long unlabelled video
Inria（法国国家信息与自动化研究所）；Univ Rennes（雷恩大学）；CNRS（法国国家科学研究中心）；IRISA（法国计算机科学与随机系统研究所）；University of Central Florida（中佛罗里达大学）；Google DeepMind（谷歌DeepMind）；University of Amsterdam（阿姆斯特丹大学）；Institute of Advanced Research on Artificial Intelligence (IARAI)（人工智能高级研究所）
King Abdullah University of Science and Technology（阿卜杜拉国王科技大学）；University of Electronic Science and Technology of China（电子科技大学）
The main question we address in this paper is how to scale up visual recognition of unseen classes, also known as zero-shot learning, to tens of thousands of categories as in the ImageNet-21K benchmark. At this scale, especially with many fine-grained categories included in ImageNet-21K, it is critical to learn quality visual semantic representations that are discriminative enough to recognize unseen classes and distinguish them from seen ones. We propose a \emph{H}ierarchical \emph{G}raphical knowledge \emph{R}epresentation framework for the confidence-based classification method, dubbed as H…
Saliency Map, the gradient of the score function with respect to the input, is the most basic technique for interpreting deep neural network decisions. However, saliency maps are often visually noisy. Although several hypotheses were proposed to account for this phenomenon, there are few works that provide rigorous analyses of noisy saliency maps. In this paper, we firstly propose a new hypothesis that noise may occur in saliency maps when irrelevant features pass through ReLU activation functions. Then, we propose Rectified Gradient, a method that alleviates this problem through layer-wise th…
Increasingly more similarities between human vision and convolutional neural networks (CNNs) have been revealed in the past few years. Yet, vanilla CNNs often fall short in generalizing to adversarial or out-of-distribution (OOD) examples which humans demonstrate superior performance. Adversarial training is a leading learning algorithm for improving the robustness of CNNs on adversarial and OOD data; however, little is known about the properties, specifically the shape bias and internal features learned inside adversarially-robust CNNs. In this paper, we perform a thorough, systematic study t…
An analysis of different techniques for recognizing and detecting objects under extreme scale variation is presented. Scale specific and scale invariant design of detectors are compared by training them with different configurations of input data. By evaluating the performance of different network architectures for classifying small objects on ImageNet, we show that CNNs are not robust to changes in scale. Based on this analysis, we propose to train and test detectors on the same scales of an image-pyramid. Since small and large objects are difficult to recognize at smaller and larger scales r…
Bimanual gestures are of the utmost importance for the study of motor coordination in humans and in everyday activities. A reliable detection of bimanual gestures in unconstrained environments is fundamental for their clinical study and to assess common activities of daily living. This paper investigates techniques for a reliable, unconstrained detection and classification of bimanual gestures. It assumes the availability of inertial data originating from the two hands/arms, builds upon a previously developed technique for gesture modelling based on Gaussian Mixture Modelling (GMM) and Gaussia…
DeepMind；University of California, Berkeley（加利福尼亚大学伯克利分校）
Human observers can learn to recognize new categories of images from a handful of examples, yet doing so with artificial ones remains an open challenge. We hypothesize that data-efficient recognition is enabled by representations which make the variability in natural signals more predictable. We therefore revisit and improve Contrastive Predictive Coding, an unsupervised objective for learning such representations. This new implementation produces features which support state-of-the-art linear classification accuracy on the ImageNet dataset. When used as input for non-linear classification wit…
The literature has proposed several methods to finetune pretrained GANs on new datasets, which typically results in higher performance compared to training from scratch, especially in the limited-data regime. However, despite the apparent empirical benefits of GAN pretraining, its inner mechanisms were not analyzed in-depth, and understanding of its role is not entirely clear. Moreover, the essential practical details, e.g., selecting a proper pretrained GAN checkpoint, currently do not have rigorous grounding and are typically determined by trial and error. This work aims to dissect the proce…
Convolutional Neural Networks demonstrate high performance on ImageNet Large-Scale Visual Recognition Challenges contest. Nevertheless, the published results only show the overall performance for all image classes. There is no further analysis why certain images get worse results and how they could be improved. In this paper, we provide deep performance analysis based on different types of images and point out the weaknesses of convolutional neural networks through experiment. We design a novel multiple paths convolutional neural network, which feeds different versions of images into separated…
Department of Computing, Imperial College London（伦敦帝国理工学院计算机系）
Deep convolution networks have proved very successful with big datasets such as the 1000-classes ImageNet. Results show that the error rate increases slowly as the size of the dataset increases. Experiments presented here may explain why these networks are very effective in solving big recognition problems. If the big task is made up of multiple smaller tasks, then the results show the ability of deep convolution networks to decompose the complex task into a number of smaller tasks and to learn them simultaneously. The results show that the performance of solving the big task on a single netwo…
Despite excellent performance on stationary test sets, deep neural networks (DNNs) can fail to generalize to out-of-distribution (OoD) inputs, including natural, non-adversarial ones, which are common in real-world settings. In this paper, we present a framework for discovering DNN failures that harnesses 3D renderers and 3D models. That is, we estimate the parameters of a 3D renderer that cause a target DNN to misbehave in response to the rendered image. Using our framework and a self-assembled dataset of 3D objects, we investigate the vulnerability of DNNs to OoD poses of well-known objects …
The state of neural network pruning has been noticed to be unclear and even confusing for a while, largely due to "a lack of standardized benchmarks and metrics" [3]. To standardize benchmarks, first, we need to answer: what kind of comparison setup is considered fair? This basic yet crucial question has barely been clarified in the community, unfortunately. Meanwhile, we observe several papers have used (severely) sub-optimal hyper-parameters in pruning experiments, while the reason behind them is also elusive. These sub-optimal hyper-parameters further exacerbate the distorted benchmarks, re…
Why Does It Look There? Structured Explanations for Image Classification
Department of Computer Science, Tulane University, New Orleans LA 70118, USA（计算机科学系，路易斯安那大学，新奥尔良 LA 70118, 美国）；Department of Biochemistry and Molecular Biology, Tulane University School of Medicine, New Orleans LA 70112, USA（生物化学与分子生物学系，路易斯安那大学医学院，新奥尔良 LA 70112, 美国）
My goal in this paper is twofold: to study how well deep models can understand the images generated by DALL-E 2 and Midjourney, and to quantitatively evaluate these generative models. Two sets of generated images are collected for object recognition and visual question answering (VQA) tasks. On object recognition, the best model, out of 10 state-of-the-art object recognition models, achieves about 60\% and 80\% top-1 and top-5 accuracy, respectively. These numbers are much lower than the best accuracy on the ImageNet dataset (91\% and 99\%). On VQA, the OFA model scores 77.3\% on answering 241…
MultiGrain is a network architecture producing compact vector representations that are suited both for image classification and particular object retrieval. It builds on a standard classification trunk. The top of the network produces an embedding containing coarse and fine-grained information, so that images can be recognized based on the object class, particular object, or if they are distorted copies. Our joint training is simple: we minimize a cross-entropy loss for classification and a ranking loss that determines if two images are identical up to data augmentation, with no need for addit…
With the success of new computational architectures for visual processing, such as convolutional neural networks (CNN) and access to image databases with millions of labeled examples (e.g., ImageNet, Places), the state of the art in computer vision is advancing rapidly. One important factor for continued progress is to understand the representations that are learned by the inner layers of these deep architectures. Here we show that object detectors emerge from training CNNs to perform scene classification. As scenes are composed of objects, the CNN for scene classification automatically discov…
Efficient Rotation Invariance in Deep Neural Networks through Artificial Mental Rotation
ZHAW Centre for Artificial Intelligence（苏黎世应用科学大学人工智能中心）；University of Lugano（卢加诺大学）；European Centre for Living Technology（欧洲生命技术中心）；The Swiss AI Lab IDSIA（瑞士人工智能实验室 IDSIA）；King Abdullah University of Science and Technology (KAUST)（阿卜杜拉国王科技大学）
We present a hierarchical VAE that, for the first time, generates samples quickly while outperforming the PixelCNN in log-likelihood on all natural image benchmarks. We begin by observing that, in theory, VAEs can actually represent autoregressive models, as well as faster, better models if they exist, when made sufficiently deep. Despite this, autoregressive models have historically outperformed VAEs in log-likelihood. We test if insufficient depth explains why by scaling a VAE to greater stochastic depth than previously explored and evaluating it CIFAR-10, ImageNet, and FFHQ. In comparison t…
University of California, Berkeley（加州大学伯克利分校）；Carnegie Mellon University（卡内基梅隆大学）；National University of Singapore（新加坡国立大学）；DeepMind（深度思维）；Adobe Research（奥多比研究院）
Recent progress in image recognition has stimulated the deployment of vision systems at an unprecedented scale. As a result, visual data are now often consumed not only by humans but also by machines. Existing image processing methods only optimize for better human perception, yet the resulting images may not be accurately recognized by machines. This can be undesirable, e.g., the images can be improperly handled by search engines or recommendation systems. In this work, we examine simple approaches to improve machine recognition of processed images: optimizing the recognition loss directly on…
Performance-optimized deep neural networks are evolving into worse models of inferotemporal visual cortex
Brown University（布朗大学）；Carney Institute for Brain Science, Brown University（布朗大学卡尼脑科学研究所）；Artificial and Natural Intelligence Toulouse Institute（图卢兹人工与自然智能研究所）；University of Pennsylvania（宾夕法尼亚大学）；Harvard Medical School（哈佛医学院）
Most recent semi-supervised deep learning (deep SSL) methods used a similar paradigm: use network predictions to update pseudo-labels and use pseudo-labels to update network parameters iteratively. However, they lack theoretical support and cannot explain why predictions are good candidates for pseudo-labels in the deep learning paradigm. In this paper, we propose a principled end-to-end framework named deep decipher (D2) for SSL. Within the D2 framework, we prove that pseudo-labels are related to network predictions by an exponential link function, which gives a theoretical support for using …
Most recent semi-supervised deep learning (deep SSL) methods used a similar paradigm: use network predictions to update pseudo-labels and use pseudo-labels to update network parameters iteratively. However, they lack theoretical support and cannot explain why predictions are good candidates for pseudo-labels. In this paper, we propose a principled end-to-end framework named deep decipher (D2) for SSL. Within the D2 framework, we prove that pseudo-labels are related to network predictions by an exponential link function, which gives a theoretical support for using predictions as pseudo-labels. …
Data61-CSIRO（Data61-联邦科学与工业研究组织）；Australian National University（澳大利亚国立大学）；University of Adelaide（阿德莱德大学）
We present You Only Cut Once (YOCO) for performing data augmentations. YOCO cuts one image into two pieces and performs data augmentations individually within each piece. Applying YOCO improves the diversity of the augmentation per sample and encourages neural networks to recognize objects from partial information. YOCO enjoys the properties of parameter-free, easy usage, and boosting almost all augmentations for free. Thorough experiments are conducted to evaluate its effectiveness. We first demonstrate that YOCO can be seamlessly applied to varying data augmentations, neural network architec…
The internal states of most deep neural networks are difficult to interpret, which makes diagnosis and debugging during training challenging. Activation maximization methods are widely used, but lead to multiple optima and are hard to interpret (appear noise-like) for complex neurons. Image-based methods use maximally-activating image regions which are easier to interpret, but do not provide pixel-level insight into why the neuron responds to them. In this work we introduce an MCMC method: Langevin Dynamics Activation Maximization (LDAM), which is designed for diagnostic visualization. LDAM pr…
The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳））
Filter pruning has attracted increasing attention in recent years for its capacity in compressing and accelerating convolutional neural networks. Various data-independent criteria, including norm-based and relationship-based ones, were proposed to prune the most unimportant filters. However, these state-of-the-art criteria fail to fully consider the dissimilarity of filters, and thus might lead to performance degradation. In this paper, we first analyze the limitation of relationship-based criteria with examples, and then introduce a new data-independent criterion, Weighted Hybrid Criterion (W…
L3i, Université de La-Rochelle（拉罗谢尔大学L3i实验室（L3i, Université de La-Rochelle））；Université de La-Rochelle（拉罗谢尔大学）
There have been many work in the literature on generation of various kinds of images such as Hand-Written characters (MNIST dataset), scene images (CIFAR-10 dataset), various objects images (ImageNet dataset), road signboard images (SVHN dataset) etc. Unfortunately, there have been very limited amount of work done in the domain of document image processing. Automatic image generation can lead to the enormous increase of labeled datasets with the help of only limited amount of labeled data. Various kinds of Deep generative models can be primarily divided into two categories. First category is a…
The field of image classification has shown an outstanding success thanks to the development of deep learning techniques. Despite the great performance obtained, most of the work has focused on natural images ignoring other domains like artistic depictions. In this paper, we use transfer learning techniques to propose a new classification network with better performance in illustration images. Starting from the deep convolutional network VGG19, pre-trained with natural images, we propose two novel models which learn object representations in the new domain. Our optimized network will learn new…
Google AI Healthcare（谷歌AI医疗）；Google Research（谷歌研究院）；Google Inc.（谷歌公司）
We study the attribution problem [28] for deep networks applied to perception tasks. For vision tasks, attribution techniques attribute the prediction of a network to the pixels of the input image. We propose a new technique called \emph{Blur Integrated Gradients}. This technique has several advantages over other methods. First, it can tell at what scale a network recognizes an object. It produces scores in the scale/frequency dimension, that we find captures interesting phenomena. Second, it satisfies the scale-space axioms [14], which imply that it employs perturbations that are free of arti…
Vision models are interpretable when they classify objects on the basis of features that a person can directly understand. Recently, methods relying on visual feature prototypes have been developed for this purpose. However, in contrast to how humans categorize objects, these approaches have not yet made use of any taxonomical organization of class labels. With such an approach, for instance, we may see why a chimpanzee is classified as a chimpanzee, but not why it was considered to be a primate or even an animal. In this work we introduce a model that uses hierarchically organized prototypes …
Chittagong University of Engineering and Technology（吉大港工程技术大学）
The Novel Coronavirus disease 2019 (COVID-19) is a fatal infectious disease, first recognized in December 2019 in Wuhan, Hubei, China, and has gone on an epidemic situation. Under these circumstances, it became more important to detect COVID-19 in infected people. Nowadays, the testing kits are gradually lessening in number compared to the number of infected population. Under recent prevailing conditions, the diagnosis of lung disease by analyzing chest CT (Computed Tomography) images has become an important tool for both diagnosis and prophecy of COVID-19 patients. In this study, a Transfer l…
The College of Management Academic Studies（管理学术学院）；The Weizmann Institute of Science（魏茨曼科学研究所）
Rich empirical evidence has shown that visual object recognition in the brain is fast and effortless, with relevant brain signals reported to start as early as 80 ms. Here we study the time trajectory of the recognition process at the level of minimal recognizable images (termed MIRC). These are images that can be recognized reliably, but in which a minute change of the image (reduction by either size or resolution) has a drastic effect on recognition. Subjects were assigned to one of nine exposure conditions: 200, 500, 1000, 2000 ms with or without masking, as well as unlimited time. The subj…
The redundancy is widely recognized in Convolutional Neural Networks (CNNs), which enables to remove unimportant filters from convolutional layers so as to slim the network with acceptable performance drop. Inspired by the linear and combinational properties of convolution, we seek to make some filters increasingly close and eventually identical for network slimming. To this end, we propose Centripetal SGD (C-SGD), a novel optimization method, which can train several filters to collapse into a single point in the parameter hyperspace. When the training is completed, the removal of the identica…
