We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

The ImageNet Large Scale Visual Recognition Challenge is a benchmark in object category classification and detection on hundreds of object categories and millions of images. The challenge has been run annually from 2010 to present, attracting participation from more than fifty institutions. This paper describes the creation of this benchmark dataset and the advances in object recognition that have been possible as a result. We discuss the challenges of collecting large-scale ground truth annotation, highlight key breakthroughs in categorical object recognition, provide a detailed analysis of the current state of the field of large-scale image classification and object detection, and compare the state-of-the-art computer vision accuracy with human accuracy. We conclude with lessons learned in the 5 years of the challenge, and propose future directions and improvements.

/pdf/imagenet-large-scale-visual-recognition-challenge-caky5nmmix.pdf

ImageNet Large Scale Visual Recognition Challenge

The central building block of convolutional neural networks (CNNs) is the convolution operator, which enables networks to construct informative features by fusing both spatial and channel-wise information within local receptive fields at each layer. A broad range of prior research has investigated the spatial component of this relationship, seeking to strengthen the representational power of a CNN by enhancing the quality of spatial encodings throughout its feature hierarchy. In this work, we focus instead on the channel relationship and propose a novel architectural unit, which we term the “Squeeze-and-Excitation” (SE) block, that adaptively recalibrates channel-wise feature responses by explicitly modelling interdependencies between channels. We show that these blocks can be stacked together to form SENet architectures that generalise extremely effectively across different datasets. We further demonstrate that SE blocks bring significant improvements in performance for existing state-of-the-art CNNs at slight additional computational cost. Squeeze-and-Excitation Networks formed the foundation of our ILSVRC 2017 classification submission which won first place and reduced the top-5 error to 2.251 percent, surpassing the winning entry of 2016 by a relative improvement of ${\sim }$ ∼ 25 percent. Models and code are available at https://github.com/hujie-frank/SENet .

/pdf/squeeze-and-excitation-networks-50icgwqgeu.pdf

Squeeze-and-Excitation Networks

Large Convolutional Network models have recently demonstrated impressive classification performance on the ImageNet benchmark Krizhevsky et al. [18]. However there is no clear understanding of why they perform so well, or how they might be improved. In this paper we explore both issues. We introduce a novel visualization technique that gives insight into the function of intermediate feature layers and the operation of the classifier. Used in a diagnostic role, these visualizations allow us to find model architectures that outperform Krizhevsky et al on the ImageNet classification benchmark. We also perform an ablation study to discover the performance contribution from different model layers. We show our ImageNet model generalizes well to other datasets: when the softmax classifier is retrained, it convincingly beats the current state-of-the-art results on Caltech-101 and Caltech-256 datasets.

https://link.springer.com/content/pdf/10.1007%2F978-3-319-10590-1_53.pdf

Visualizing and Understanding Convolutional Networks

In this work we address the task of semantic image segmentation with Deep Learning and make three main contributions that are experimentally shown to have substantial practical merit. First , we highlight convolution with upsampled filters, or ‘atrous convolution’, as a powerful tool in dense prediction tasks. Atrous convolution allows us to explicitly control the resolution at which feature responses are computed within Deep Convolutional Neural Networks. It also allows us to effectively enlarge the field of view of filters to incorporate larger context without increasing the number of parameters or the amount of computation. Second , we propose atrous spatial pyramid pooling (ASPP) to robustly segment objects at multiple scales. ASPP probes an incoming convolutional feature layer with filters at multiple sampling rates and effective fields-of-views, thus capturing objects as well as image context at multiple scales. Third , we improve the localization of object boundaries by combining methods from DCNNs and probabilistic graphical models. The commonly deployed combination of max-pooling and downsampling in DCNNs achieves invariance but has a toll on localization accuracy. We overcome this by combining the responses at the final DCNN layer with a fully connected Conditional Random Field (CRF), which is shown both qualitatively and quantitatively to improve localization performance. Our proposed “DeepLab” system sets the new state-of-art at the PASCAL VOC-2012 semantic image segmentation task, reaching 79.7 percent mIOU in the test set, and advances the results on three other datasets: PASCAL-Context, PASCAL-Person-Part, and Cityscapes. All of our code is made publicly available online.

DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs

Image classification techniques using images with separate grayscale and color channels are described. In one or more implementations, an image classification network includes grayscale filters and color filters which are separate from the grayscale filters. The grayscale filters are configured to extract grayscale features from a grayscale channel of an image, and the color filters are configured to extract color features from a color channel of the image. The extracted grayscale features and color features are used to identify an object in the image, and the image is classified based on the identified object.

Image classification using images with separate grayscale and color channels

In recent years, designing the coding and pooling structures in layered networks has been shown to be a useful method for learning high-level feature representations for visual data. Yet, such learning structures have not been extensively studied for audio signals. In this paper, we investigate different pooling strategies based on the sparse coding scheme and propose a temporal pyramid pooling method to extract discriminative and shiftinvariant feature representations. We demonstrate the superiority of our new feature representation over traditional features on the acoustic event classification task.

http://www.ifp.illinois.edu/~huang146/papers/Pooling_Spikegram_interspeech2012.pdf

Pooling robust shift-invariant sparse representations of acoustic signals

Action recognition with 3D skeleton sequences is becoming popular due to its speed and robustness. The recently proposed Convolutional Neural Networks (CNN) based methods have shown good performance in learning spatio-temporal representations for skeleton sequences. Despite the good recognition accuracy achieved by previous CNN based methods, there exist two problems that potentially limit the performance. First, previous skeleton representations are generated by chaining joints with a fixed order. The corresponding semantic meaning is unclear and the structural information among the joints is lost. Second, previous models do not have an ability to focus on informative joints. The attention mechanism is important for skeleton based action recognition because there exist spatio-temporal key stages and the joint predictions can be inaccurate. To solve the two problems, we propose a novel CNN based method for skeleton based action recognition. We first redesign the skeleton representations with a depth-first tree traversal order, which enhances the semantic meaning of skeleton images and better preserves the structural information. We then propose the idea of a two-branch attention architecture that focuses on spatio-temporal key stages and filters out unreliable joint predictions. A base attention model with the simplest structure is first introduced to illustrate the two-branch attention architecture. By improving the structures in both branches, we further propose a Global Long-sequence Attention Network (GLAN). Experiment results on the NTU RGB+D dataset and the SBU Kinetic Interaction dataset show that our proposed approach outperforms the state-of-the-art, as well as the effectiveness of each component.

Action Recognition with Visual Attention on Skeleton Images

Embodiments of the present invention relate to finding semantic parts in images. In implementation, a convolutional neural network (CNN) is applied to a set of images to extract features for each image. Each feature is defined by a feature vector that enables a subset of the set of images to be clustered in accordance with a similarity between feature vectors. Normalized cuts may be utilized to help preserve pose within each cluster. The images in the cluster are aligned and part proposals are generated by sampling various regions in various sizes across the aligned images. To determine which part proposal corresponds to a semantic part, a classifier is trained for each part proposal and semantic part to determine which part proposal best fits the correlation pattern given by the true semantic part. In this way, semantic parts in images can be identified without any previous part annotations.

Finding semantic parts in images

Image priors are essential to many image restoration applications, including denoising, deblurring, and inpainting. Existing methods use either priors from the given image (internal) or priors from a separate collection of images (external). We find through statistical analysis that unifying the internal and external patch priors may yield a better patch prior. We propose a novel prior learning algorithm that combines the strength of both internal and external priors. In particular, we first learn a generic Gaussian mixture model from a collection of training images and then adapt the model to the given image by simultaneously adding additional components and refining the component parameters. We apply this image-specific prior to image denoising. The experimental results show that our approach yields better or competitive denoising results in terms of both the peak signal-to-noise ratio and structural similarity.

Jianchao Yang

Papers

Image classification using images with separate grayscale and color channels

Pooling robust shift-invariant sparse representations of acoustic signals

Action Recognition with Visual Attention on Skeleton Images

Finding semantic parts in images

Image-Specific Prior Adaptation for Denoising