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.

Deep Lip Reading: a comparison of models and an online application

Abstract: The goal of this paper is to develop state-of-the-art models for lip reading -- visual speech recognition. We develop three architectures and compare their accuracy and training times: (i) a recurrent model using LSTMs; (ii) a fully convolutional model; and (iii) the recently proposed transformer model. The recurrent and fully convolutional models are trained with a Connectionist Temporal Classification loss and use an explicit language model for decoding, the transformer is a sequence-to-sequence model. Our best performing model improves the state-of-the-art word error rate on the challenging BBC-Oxford Lip Reading Sentences 2 (LRS2) benchmark dataset by over 20 percent. As a further contribution we investigate the fully convolutional model when used for online (real time) lip reading of continuous speech, and show that it achieves high performance with low latency.

Amplifying Key Cues for Human-Object-Interaction Detection

Abstract: Human-object interaction (HOI) detection aims to detect and recognise how people interact with the objects that surround them. This is challenging as different interaction categories are often distinguished only by very subtle visual differences in the scene. In this paper we introduce two methods to amplify key cues in the image, and also a method to combine these and other cues when considering the interaction between a human and an object. First, we introduce an encoding mechanism for representing the fine-grained spatial layout of the human and object (a subtle cue) and also semantic context (a cue, represented by text embeddings of surrounding objects). Second, we use plausible future movements of humans and objects as a cue to constrain the space of possible interactions. Third, we use a gate and memory architecture as a fusion module to combine the cues. We demonstrate that these three improvements lead to a performance which exceeds prior HOI methods across standard benchmarks by a considerable margin.

Euclidean structure from uncalibrated images

Abstract: A number of recent papers have demonstrated that camera “self-calibration” can be accomplished purely from image measurements, without requiring special calibration objects or known camera motion. We describe a method, based on self-calibration, for obtaining (scaled) Euclidean structure from multiple uncalibrated perspective images using only point matches between views. The method is in two stages. First, using an uncalibrated camera, structure is recovered up to an affine ambiguity from two views. Second, from one or more further views of this affine structure the camera intrinsic parameters are determined, and the structure ambiguity reduced to scaled Euclidean. The technique is independent of how the affine structure is obtained. We analyse its limitations and degeneracies. Results are given for images of real scenes. An application is described for active vision, where a Euclidean reconstruction is obtained during normal operation with an initially uncalibrated camera. Finally, it is demonstrated that Euclidean reconstruction can be obtained from a single perspective image of a repeated structure

Talking Heads: Detecting Humans and Recognizing Their Interactions

Abstract: The objective of this work is to accurately and efficiently detect configurations of one or more people in edited TV material. Such configurations often appear in standard arrangements due to cinematic style, and we take advantage of this to provide scene context. We make the following contributions: first, we introduce a new learnable context aware configuration model for detecting sets of people in TV material that predicts the scale and location of each upper body in the configuration, second, we show that inference of the model can be solved globally and efficiently using dynamic programming, and implement a maximum margin learning framework, and third, we show that the configuration model substantially outperforms a Deformable Part Model (DPM) for predicting upper body locations in video frames, even when the DPM is equipped with the context of other upper bodies. Experiments are performed over two datasets: the TV Human Interaction dataset, and 150 episodes from four different TV shows. We also demonstrate the benefits of the model in recognizing interactions in TV shows.

Motion From Point Matches Using Affine Epipolar Geometry

Abstract: Algorithms to perform point-based motion estimation under orthographic and scaled orthographic projection abound in the literature. A key limitation of many existing algorithms is that they rely on the selection of a minimal point set to define a “local coordinate frame”. This approach is extremely sensitive to errors and noise, and forfeits the advantages of using the full data set. Furthermore, attention is seldom paid to the statistical performance of the algorithms. We present a new framework that caters for errors and noise, and allows all available features to be used, without the need to select a frame explicitly. This theory is derived in the context of the affine camera, which generalises the orthographic, scaled orthographic and para-perspective models. We define the affine epipolar geometry for two such cameras, giving the fundamental matrix in this case and discussing its noise resistant computation. The two-view rigid motion parameters (the scale factor between views, projection of the 3D axis of rotation and cyclotorsion angle) are then determined directly from the epipolar geometry. Optimal estimates are obtained over time by means of a linear Kalman filter, and results are presented on real data.

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 .