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

The paper has two main contributions: The first is a set of methods for computing structure and motion for m ≥ 3 views of 6 points. It is shown that a geometric image error can be minimized over all views by a simple three parameter numerical optimization. Then, that an algebraic image error can be minimized over all views by computing the solution to a cubic in one variable. Finally, a minor point, is that this "quasi-linear" linear solution enables a more concise algorithm, than any given previously, for the reconstruction of 6 points in 3 views.

The second contribution is an m view n ≥ 6 point robust reconstruction algorithm which uses the 6 point method as a search engine. This extends the successful RANSAC based algorithms for 2-views and 3-views to m views. The algorithm can cope with missing data and mismatched data and may be used as an efficient initializer for bundle adjustment.

The new algorithms are evaluated on synthetic and real image sequences, and compared to optimal estimation results (bundle adjustment).

/pdf/a-six-point-solution-for-structure-and-motion-1ugnq16hb8.pdf

A Six Point Solution for Structure and Motion

We propose a self-supervised framework for learning facial attributes by simply watching videos of a human face speaking, laughing, and moving over time. To perform this task, we introduce a network, Facial Attributes-Net (FAb-Net), that is trained to embed multiple frames from the same video face-track into a common low-dimensional space. With this approach, we make three contributions: first, we show that the network can leverage information from multiple source frames by predicting confidence/attention masks for each frame; second, we demonstrate that using a curriculum learning regime improves the learned embedding; finally, we demonstrate that the network learns a meaningful face embedding that encodes information about head pose, facial landmarks and facial expression, i.e. facial attributes, without having been supervised with any labelled data. We are comparable or superior to state-of-the-art self-supervised methods on these tasks and approach the performance of supervised methods.

Self-supervised learning of a facial attribute embedding from video.

Self-supervised learning of a facial attribute embedding from video

We study the problem of matching photos of a person to paintings of that person, in order to retrieve similar paintings given a query photo. This is challenging as paintings span many media (oil, ink, watercolor) and can vary tremendously in style (caricature, pop art, minimalist). We make the following contributions: (i) we show that, depending on the face representation used, performance can be improved substantially by learning – either by a linear projection matrix common across identities, or by a per-identity classifier. We compare Fisher Vector and Convolutional Neural Network representations for this task; (ii) we introduce new datasets for learning and evaluating this problem; (iii) we also consider the reverse problem of retrieving photos from a large corpus given a painting; and finally, (iv) using the learnt descriptors, we show that, given a photo of a person, we are able to find their doppelganger in a large dataset of oil paintings, and how this result can be varied by modifying attributes (e.g. frowning, old looking).

/pdf/face-painting-querying-art-with-photos-132jfzhudq.pdf

Face Painting: querying art with photos

The goal of this work is to automatically learn a large number of signs from sign language-interpreted TV broadcasts. We achieve this by exploiting supervisory information available in the subtitles of the broadcasts. However, this information is both weak and noisy and this leads to a challenging correspondence problem when trying to identify the temporal window of the sign. We make the following contributions: (i) we show that, somewhat counter-intuitively, mouth patterns are highly informative for isolating words in a language for the Deaf, and their co-occurrence with signing can be used to significantly reduce the correspondence search space; and (ii) we develop a multiple instance learning method using an efficient discriminative search, which determines a candidate list for the sign with both high recall and precision. We demonstrate the method on videos from BBC TV broadcasts, and achieve higher accuracy and recall than previous methods, despite using much simpler features.

/pdf/large-scale-learning-of-sign-language-by-watching-tv-using-3fb3295std.pdf

Andrew Zisserman

Papers

A Six Point Solution for Structure and Motion

Self-supervised learning of a facial attribute embedding from video.

Self-supervised learning of a facial attribute embedding from video

Face Painting: querying art with photos

Large-scale Learning of Sign Language by Watching TV (Using Co-occurrences).