Deeper neural networks are more difficult to train. We present a residual learning framework to ease the training of networks that are substantially deeper than those used previously. We explicitly reformulate the layers as learning residual functions with reference to the layer inputs, instead of learning unreferenced functions. We provide comprehensive empirical evidence showing that these residual networks are easier to optimize, and can gain accuracy from considerably increased depth. On the ImageNet dataset we evaluate residual nets with a depth of up to 152 layers—8× deeper than VGG nets [40] but still having lower complexity. An ensemble of these residual nets achieves 3.57% error on the ImageNet test set. This result won the 1st place on the ILSVRC 2015 classification task. We also present analysis on CIFAR-10 with 100 and 1000 layers. The depth of representations is of central importance for many visual recognition tasks. Solely due to our extremely deep representations, we obtain a 28% relative improvement on the COCO object detection dataset. Deep residual nets are foundations of our submissions to ILSVRC & COCO 2015 competitions1, where we also won the 1st places on the tasks of ImageNet detection, ImageNet localization, COCO detection, and COCO segmentation.

/pdf/deep-residual-learning-for-image-recognition-b75jg61qrq.pdf

Deep Residual Learning for Image Recognition

In this work we investigate the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting. Our main contribution is a thorough evaluation of networks of increasing depth using an architecture with very small (3x3) convolution filters, which shows that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 weight layers. These findings were the basis of our ImageNet Challenge 2014 submission, where our team secured the first and the second places in the localisation and classification tracks respectively. We also show that our representations generalise well to other datasets, where they achieve state-of-the-art results. We have made our two best-performing ConvNet models publicly available to facilitate further research on the use of deep visual representations in computer vision.

Very Deep Convolutional Networks for Large-Scale Image Recognition

The explosion of image data on the Internet has the potential to foster more sophisticated and robust models and algorithms to index, retrieve, organize and interact with images and multimedia data. But exactly how such data can be harnessed and organized remains a critical problem. We introduce here a new database called “ImageNet”, a large-scale ontology of images built upon the backbone of the WordNet structure. ImageNet aims to populate the majority of the 80,000 synsets of WordNet with an average of 500-1000 clean and full resolution images. This will result in tens of millions of annotated images organized by the semantic hierarchy of WordNet. This paper offers a detailed analysis of ImageNet in its current state: 12 subtrees with 5247 synsets and 3.2 million images in total. We show that ImageNet is much larger in scale and diversity and much more accurate than the current image datasets. Constructing such a large-scale database is a challenging task. We describe the data collection scheme with Amazon Mechanical Turk. Lastly, we illustrate the usefulness of ImageNet through three simple applications in object recognition, image classification and automatic object clustering. We hope that the scale, accuracy, diversity and hierarchical structure of ImageNet can offer unparalleled opportunities to researchers in the computer vision community and beyond.

/pdf/imagenet-a-large-scale-hierarchical-image-database-1e8f663s1h.pdf

ImageNet: A large-scale hierarchical image database

There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net .

/pdf/u-net-convolutional-networks-for-biomedical-image-2xoyjzggxv.pdf

U-Net: Convolutional Networks for Biomedical Image Segmentation

Recent progress in fine-grained gesture and action classification, and machine translation, point to the possibility of automated sign language recognition becoming a reality. A key stumbling block in making progress towards this goal is a lack of appropriate training data, stemming from the high complexity of sign annotation and a limited supply of qualified annotators. In this work, we introduce a new scalable approach to data collection for sign recognition in continuous videos. We make use of weakly-aligned subtitles for broadcast footage together with a keyword spotting method to automatically localise sign-instances for a vocabulary of 1,000 signs in 1,000 h of video. We make the following contributions: (1) We show how to use mouthing cues from signers to obtain high-quality annotations from video data—the result is the BSL-1K dataset, a collection of British Sign Language (BSL) signs of unprecedented scale; (2) We show that we can use BSL-1K to train strong sign recognition models for co-articulated signs in BSL and that these models additionally form excellent pretraining for other sign languages and benchmarks—we exceed the state of the art on both the MSASL and WLASL benchmarks. Finally, (3) we propose new large-scale evaluation sets for the tasks of sign recognition and sign spotting and provide baselines which we hope will serve to stimulate research in this area.

BSL-1K: Scaling Up Co-articulated Sign Language Recognition Using Mouthing Cues

New methods are reported for the detection of multiple solutions (degeneracy) when estimating the fundamental matrix, with specific emphasis on robustness in the presence of data contamination (outliers). The fundamental matrix can be used as a first step in the recovery of structure from motion. If the set of correspondences is degenerate then this structure cannot be accurately recovered and many solutions will explain the data equally well. It is essential that we are alerted to such eventualities. However, current feature matchers are very prone to mismatching, giving a high rate of contamination within the data. Such contamination can make a degenerate data set appear non degenerate, thus the need for robust methods becomes apparent. The paper presents such methods with a particular emphasis on providing a method that will work on real imagery and with an automated (non perfect) feature detector and matcher. It is demonstrated that proper modelling of degeneracy in the presence of outliers enables the detection of outliers which would otherwise be missed. Results using real image sequences are presented. All processing, point matching, degeneracy detection and outlier detection is automatic. >

https://www.computer.org/csdl/pds/api/csdl/proceedings/download-article/12OmNyqRniE/pdf

Robust detection of degenerate configurations for the fundamental matrix

We propose two complementary techniques to improve the performance of action recognition systems. The first technique addresses the temporal interval ambiguity of actions by learning a classifier score distribution over video subsequences. A classifier based on this score distribution is shown to be more effective than using the maximum or average scores. The second technique learns a classifier for the relative values of action scores, capturing the correlation and exclusion between action classes. Both techniques are simple and have efficient implementations using a Least-Squares SVM. We demonstrate that taken together the techniques exceed the state-of-the-art performance by a wide margin on challenging benchmarks for human actions.

/pdf/improving-human-action-recognition-using-score-distribution-46scx0hkyj.pdf

Improving Human Action Recognition Using Score Distribution and Ranking

This paper presents all algorithm for computing a maximum likelihood estimate (MLE) of the trifocal tensor. The input to the algorithm is three images of the same scene, and the output is the estimated tensor and corner and line feature matches across the three images that are consistent with this estimate. Particular novelties of the algorithm are the computation of a trifocal tensor from six point correspondences, and a parameterization of the trifocal tensor which enforces the constraints between the tensor elements. The algorithm uses techniques from robust statistics and is fully automatic. Results are presented for synthetic and real image triplets. The proposed parameterization is compared to other existing methods.

Robust Parameterization and Computation of the Trifocal Tensor.

Kernel methods yield state-of-the-art performance in certain applications such as image classification and object detection. However, large scale problems require machine learning techniques of at most linear complexity and these are usually limited to linear kernels. This unfortunately rules out gold-standard kernels such as the generalized RBF kernels (e.g. exponential-c 2 ). Recently, Maji and Berg [13] and Vedaldi and Zisserman [20] proposed explicit feature maps to approximate the additive kernels (intersection, c 2 , etc.) by linear ones, thus enabling the use of fast machine learning technique in a non-linear context. An analogous technique was proposed by Rahimi and Recht [14] for the translation invariant RBF kernels. In this paper, we complete the construction and combine the two techniques to obtain explicit feature maps for the generalized RBF kernels. Furthermore, we investigate a learning method using l 1 regularization to encourage sparsity in the final vector representation, and thus reduce its dimension. We evaluate this technique on the VOC 2007 detection challenge, showing when it can improve on fast additive kernels, and the trade-offs in complexity and accuracy.

/pdf/generalized-rbf-feature-maps-for-efficient-detection-25eir8o9l6.pdf

Andrew Zisserman

Papers

BSL-1K: Scaling Up Co-articulated Sign Language Recognition Using Mouthing Cues

Robust detection of degenerate configurations for the fundamental matrix

Improving Human Action Recognition Using Score Distribution and Ranking

Robust Parameterization and Computation of the Trifocal Tensor.

Generalized RBF feature maps for Efficient Detection