Andrew Zisserman

Bio: Andrew Zisserman is an academic researcher from University of Oxford. The author has contributed to research in topics: Real image & Convolutional neural network. The author has an hindex of 167, co-authored 808 publications receiving 261717 citations. Previous affiliations of Andrew Zisserman include University of Edinburgh & Microsoft.

Real-time Visual Tracking for Surveillance and Path Planning

Abstract: In this paper we report progress towards a flexible, visually driven, object manipulation system. The aim is that a robot arm with a camera and gripper mounted on its tip should be able to transport objects across an obstacle-strewn environment. Our system is based on the analysis of moving image contours, which can provide direct estimates of the shape of curved surfaces. Recently we have elaborated on this basis in two respects. First we have developed real-time visual tracking methods using “dynamic contours” with Lagrangian Dynamics allowing direct generation of approximations to geodesic paths around obstacles. Secondly we have built a 2 1/2D system for incremental, active exploration of free-space.

Vision algorithms : theory and practice : International Workshop on Vision Algorithms, Corfu, Greece, September 21-22, 1999 : proceedings

Abstract: Correspondence and Tracking.- An Experimental Comparison of Stereo Algorithms.- A General Method for Feature Matching and Model Extraction.- Characterizing the Performance of Multiple-Image Point-Correspondence Algorithms Using Self-Consistency.- A Sampling Algorithm for Tracking Multiple Objects.- Real-Time Tracking of Complex Structures for Visual Servoing.- Geometry and Reconstruction.- Direct Recovery of Planar-Parallax from Multiple Frames.- Generalized Voxel Coloring.- Projective Reconstruction from N Views Having One View in Common.- Point- and Line-Based Parameterized Image Varieties for Image-Based Rendering.- Recovery of Circular Motion from Profiles of Surfaces.- Optimal Reconstruction.- Optimization Criteria, Sensitivity and Robustness of Motion and Structure Estimation.- Gauge Independence in Optimization Algorithms for 3D Vision.- Uncertainty Modeling for Optimal Structure from Motion.- Error Characterization of the Factorization Approach to Shape and Motion Recovery.- Bootstrapping Errors-in-Variables Models.- Invited Talks.- Annotation of Video by Alignment to Reference Imagery.- Computer-Vision for the Post-production World: Facts and Challenges through the REALViZ Experience.- Special Sessions.- About Direct Methods.- Feature Based Methods for Structure and Motion Estimation.- Discussion for Direct versus Features Session.- Bundle Adjustment - A Modern Synthesis.- Discussion for Session on Bundle Adjustment.- Summary of the Panel Session.

Watch, Read and Lookup: Learning to Spot Signs from Multiple Supervisors

Abstract: The focus of this work is sign spotting—given a video of an isolated sign, our task is to identify whether and where it has been signed in a continuous, co-articulated sign language video. To achieve this sign spotting task, we train a model using multiple types of available supervision by: (1) watching existing sparsely labelled footage; (2) reading associated subtitles (readily available translations of the signed content) which provide additional weak-supervision; (3) looking up words (for which no co-articulated labelled examples are available) in visual sign language dictionaries to enable novel sign spotting. These three tasks are integrated into a unified learning framework using the principles of Noise Contrastive Estimation and Multiple Instance Learning. We validate the effectiveness of our approach on low-shot sign spotting benchmarks. In addition, we contribute a machine-readable British Sign Language (BSL) dictionary dataset of isolated signs, BslDict, to facilitate study of this task. The dataset, models and code are available at our project page (https://www.robots.ox.ac.uk/~vgg/research/bsldict/).

Action Recognition From Weak Alignment of Body Parts.

Abstract: We propose a method for human action recognition from still images that uses the silhouette and the upper body as a proxy for the pose of the person, and also to guide alignment between instances for the purpose of computing registered feature descriptors. Our contributions include an efficient algorithm, formulated as an energy minimization, for using the silhouette to align body parts between imaged human samples. The descriptors computed over the aligned body parts are incorporated in a multiple kernel framework to learn a classifier for each action class. Experiments on the challenging PASCAL VOC 2012 dataset show that our method outperforms the state-of-the-art on the majority of action classes.

Deep Residual Learning for Image Recognition

Abstract: 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.

Very Deep Convolutional Networks for Large-Scale Image Recognition

Abstract: 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

Abstract: 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.

ImageNet: A large-scale hierarchical image database

Abstract: 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.

U-Net: Convolutional Networks for Biomedical Image Segmentation

Abstract: 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 .