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

In this paper, we describe a new approach for recovering 3D geometry from an uncalibrated image sequence of a single axis (turntable) motion. Unlike previous methods, the computation of multiple views encoded by the fundamental matrix or trifocal tensor is not required. Instead, the new approach is based on fitting a conic locus to corresponding image points over multiple views. It is then shown that the geometry of single axis motion can be recovered given at least two such conics. In the case of two conics the reconstruction may have a two fold ambiguity, but this ambiguity is removed if three conics are used.The approach enables the geometry of the single axis motion (the 3D rotation axis and Euclidean geometry in planes perpendicular to this axis) to be estimated using the minimal number of parameters. It is demonstrated that a Maximum Likelihood Estimation results in measurements that are as good as or superior to those obtained by previous methods, and with a far simpler algorithm. Examples are given on various real sequences, which show the accuracy and robustness of the new algorithm.

/pdf/single-axis-geometry-by-fitting-conics-4vjlufb5am.pdf

Single Axis Geometry by Fitting Conics

The objective of this work is to learn sub-categories. Rather than casting this as a problem of unsupervised clustering, we investigate a weakly supervised approach using both positive and negative samples of the category. We make the following contributions: (i) we introduce a new model for discriminative sub-categorization which determines cluster membership for positive samples whilst simultaneously learning a max-margin classifier to separate each cluster from the negative samples, (ii) we show that this model does not suffer from the degenerate cluster problem that afflicts several competing methods (e.g., Latent SVM and Max-Margin Clustering), (iii) we show that the method is able to discover interpretable sub-categories in various datasets. The model is evaluated experimentally over various datasets, and its performance advantages over k-means and Latent SVM are demonstrated. We also stress test the model and show its resilience in discovering sub-categories as the parameters are varied.

/pdf/discriminative-sub-categorization-1dyrm2h572.pdf

Discriminative Sub-categorization

We describe a method to automatically predict radiological scores in spinal Magnetic Resonance Images (MRIs). Furthermore, we also identify and localize the pathologies that are the reasons for these scores. We term these pathological regions the “evidence hotspots”. Our contributions are two-fold: (i) a Convolutional Neural Network (CNN) architecture and training scheme to predict multiple radiological scores on multiple slice sagittal MRIs. The scheme uses multi-task CNN training with augmentation, and handles the class imbalance common in medical classification tasks. (ii) the prediction of a heat-map of evidence hotspots for each score. For both of these, all that is required for training is the class label of the disc or vertebrae, no stronger supervision (such as slice labels) is needed. We report state-of-the-art and near-human performances across multiple radiological scorings including: Pfirrmann grading, disc narrowing, endplate defects, and marrow changes.

/pdf/spinenet-automatically-pinpointing-classification-evidence-2p6yfy6hq1.pdf

SpineNet: Automatically Pinpointing Classification Evidence in Spinal MRIs

This paper describes a method for camera calibration. The system consists of a static camera which takes a sequence of images of a calibration plane rotating around a fixed axis. There is no requirement for any exact positioning of the camera or calibration plane.

/pdf/camera-calibration-using-multiple-images-1z3rb6t4dz.pdf

Camera Calibration Using Multiple Images

Navigating through unstructured environments is a basic capability of intelligent creatures, and thus is of fundamental interest in the study and development of artificial intelligence. Long-range navigation is a complex cognitive task that relies on developing an internal representation of space, grounded by recognisable landmarks and robust visual processing, that can simultaneously support continuous self-localisation ("I am here") and a representation of the goal ("I am going there"). Building upon recent research that applies deep reinforcement learning to maze navigation problems, we present an end-to-end deep reinforcement learning approach that can be applied on a city scale. Recognising that successful navigation relies on integration of general policies with locale-specific knowledge, we propose a dual pathway architecture that allows locale-specific features to be encapsulated, while still enabling transfer to multiple cities. We present an interactive navigation environment that uses Google StreetView for its photographic content and worldwide coverage, and demonstrate that our learning method allows agents to learn to navigate multiple cities and to traverse to target destinations that may be kilometres away. The project webpage this http URL contains a video summarising our research and showing the trained agent in diverse city environments and on the transfer task, the form to request the StreetLearn dataset and links to further resources. The StreetLearn environment code is available at this https URL

Andrew Zisserman

Papers

Single Axis Geometry by Fitting Conics

Discriminative Sub-categorization

SpineNet: Automatically Pinpointing Classification Evidence in Spinal MRIs

Camera Calibration Using Multiple Images

Learning to Navigate in Cities Without a Map