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

Our goal in this paper is to investigate properties of 3D shape that can be determined from a single image. We define 3D shape attributes -- generic properties of the shape that capture curvature, contact and occupied space. Our first objective is to infer these 3D shape attributes from a single image. A second objective is to infer a 3D shape embedding -- a low dimensional vector representing the 3D shape. 
We study how the 3D shape attributes and embedding can be obtained from a single image by training a Convolutional Neural Network (CNN) for this task. We start with synthetic images so that the contribution of various cues and nuisance parameters can be controlled. We then turn to real images and introduce a large scale image dataset of sculptures containing 143K images covering 2197 works from 242 artists. 
For the CNN trained on the sculpture dataset we show the following: (i) which regions of the imaged sculpture are used by the CNN to infer the 3D shape attributes; (ii) that the shape embedding can be used to match previously unseen sculptures largely independent of viewpoint; and (iii) that the 3D attributes generalize to images of other (non-sculpture) object classes.

From Images to 3D Shape Attributes

Continuity errors occur in many movies and television series, in spite of careful checking by those employed to minimise them. In this work we develop a scheme for automatically detecting these errors and producing a ranked list of the most likely inconsistencies, working from a commercial DVD release. We use the editing structure of the movie to detect pairs of shots within a scene that might have arisen from different takes, and thus are possible candidates for continuity errors. These pairs are then registered and examined for differences that lack an obvious cause -- suppressing changes arising from humans and other moving objects by using upper body detectors and trackers. The result is a ranked list of possible continuity errors for the movie.We show discovered errors for a number of feature length movies including relatively recent releases, such as 'Love Actually' (Curtis, 2003) and classics, such as 'Pretty Woman' (Marshall, 1990). We discover mistakes that have previously been missed in the listings for these movies on such websites as moviemistakes.com.

/pdf/automatic-retrieval-of-visual-continuity-errors-in-movies-2vlrclzvqy.pdf

Automatic retrieval of visual continuity errors in movies

Although there have been some recent advances in sign language recognition, part of the problem is that most
computer scientists in this research area do not have the required in-depth knowledge of sign language, and
often have no connection with the Deaf community or sign linguists. For example, one project described as
translation into sign language aimed to take subtitles and turn them into fingerspelling. This is one of many
reasons why much of this technology, including sign-language gloves, simply doesn’t address the
challenges. However there are benefits to achieving automatic sign language recognition. The process of annotating and
analysing sign language data on video is extremely labour-intensive. Sign language recognition technology
could help speed this up.
Until recently we have lacked large signed video datasets that have been precisely and consistently
transcribed and translated – these are needed to train computers for automation. But sign language corpora –
large datasets like the British Sign Language Corpus (Schembri et al., 2014) - bring new possibilities for this
technology. Here we describe the project “ExTOL: End to End Translation of British Sign Language” – which has one
aim of building the world's first British Sign Language to English translation system and the first practically
functional machine translation system for any sign language. Annotation work previously done on the BSL
Corpus is providing essential data to be used by computer vision tools to assist with automatic recognition.
To achieve this the computer must be able to recognise not only the shape, motion and location of the hands
but also nonmanual features – including facial expression, mouth movements, and body posture of the signer.
It must also understand how all of this activity in connected signing can be translated into written/spoken language. The technology for recognising hand, face and body positions and movements is improving all the
time, which will allow significant progress in speeding up automatic recognition and identification of these
elements (e.g. recognising specific facial expressions or mouth movements or head movements). Full
translation from BSL to English is of course more complex but the automatic recognition of basic position
and movements will help pave the way towards automatic translation. In addition, a secondary aim of the
project is to create automatic annotation tools to be integrated into the software annotation package ELAN.
We will additionally make the software tools available as independent packages, thus potentially allowing
their inclusion into other annotation software such as iLex. In this poster we report on some initial progress on the ExTOL project. This includes (1) automatic
recognition of English mouthings, which is being trained using 600+ hours of audiovisual spoken English
from British television and TED videos, and developed by testing on English mouthing annotations from the
BSL Corpus. It also includes (2) translation of IDGloss to Free Translation, where the aim is to produce
English-like sentences given sign glosses in BSL word order. We report baseline results on a subset of the
BSL Corpus, which contains 10,000+ sequences and over 5,000 unique tokens, using the state-of-the-art
attention based Neural Machine Translation approaches (Camgoz et al., 2018; Vaswani et al., 2017).
Although it is clear that free translation (i.e. full English translation) cannot be achieved via ID glosses alone, this baseline translation task will help contribute to the overall BSL to English translation process - at least at the level of manual signs.

ExTOL: Automatic recognition of British Sign Language using the BSL Corpus

We describe a method to automatically predict scoliosis in Dual-energy X-ray Absorptiometry (DXA) scans. We also show that intermediate representations, which in our case are segments of body parts, help improve performance. Hence, we propose a two step process for prediction: (i) we learn to segment body parts via a segmentation Convolutional Neural Network (CNN), which we show outperforms the noisy labels it was trained on, and (ii) we predict with a classification CNN that uses as input both the raw DXA scan and also the intermediate representation, i.e. the segmented body parts. We demonstrate that this two step process can predict scoliosis with high accuracy, and can also localize the spinal curves (i.e. geometry) without additional supervision. Furthermore, we also propose a soft score of scoliosis based on the classification CNN which correlates to the severity of scoliosis.

/pdf/predicting-scoliosis-in-dxa-scans-using-intermediate-2i0g2sa2bn.pdf

Predicting Scoliosis in DXA Scans Using Intermediate Representations.

The objectives of the work described in this paper are simply stated: given examples of a particular person and an unlabelled video, we wish to find every instance of that person in the video and in others. This is an extremely difficult problem because of the many sources of variation in the person's appearance. We present a two stage approach. A 3-D ellipsoid approximation of the person's head is used to train a set of generative parts-based 'constellation' models which propose candidate detections in an image. The detected parts are then used to align the model, and the detections verified by global appearance. Novel aspects of the approach include the minimal supervision required and the generalization across a wide range of pose. We demonstrate results of detecting three characters in a TV situation comedy.

/pdf/automated-visual-identification-of-characters-in-situation-3ecsond8oy.pdf

Andrew Zisserman

Papers

From Images to 3D Shape Attributes

Automatic retrieval of visual continuity errors in movies

ExTOL: Automatic recognition of British Sign Language using the BSL Corpus

Predicting Scoliosis in DXA Scans Using Intermediate Representations.

Automated visual identification of characters in situation comedies