scispace - formally typeset
Search or ask a question
Author

Andrew Zisserman

Other affiliations: University of Edinburgh, Microsoft, University of Leeds  ...read more
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.


Papers
More filters
Book ChapterDOI
01 Mar 2004
TL;DR: In this paper, the trifocal tensor is introduced to handle the transfer from two views to a third view, where the camera matrices can be retrieved from the tensor up to a common projective transformation of 3-space, and the fundamental matrices for view-pairs may be retrieved uniquely.
Abstract: Outline This part contains two chapters on the geometry of three-views. The scene is imaged with three cameras perhaps simultaneously in a trinocular rig, or sequentially from a moving camera. Chapter 15 introduces a new multiple view object – the trifocal tensor. This has analogous properties to the fundamental matrix of two-view geometry: it is independent of scene structure depending only on the (projective) relations between the cameras. The camera matrices may be retrieved from the trifocal tensor up to a common projective transformation of 3-space, and the fundamental matrices for view-pairs may be retrieved uniquely. The new geometry compared with the two-view case is the ability to transfer from two views to a third: given a point correspondence over two views the position of the point in the third view is determined; and similarly, given a line correspondence over two views the position of the line in the third view is determined. This transfer property is of great benefit when establishing correspondences over multiple views. If the essence of the epipolar constraint over two views is that rays back-projected from corresponding points are coplanar, then the essence of the trifocal constraint over three views is the geometry of a point–line–line correspondence arising from the image of a point on a line in 3-space: corresponding image lines in two views back-project to planes which intersect in a line in 3-space, and the ray back-projected from a corresponding image point in a third view must intersect this line.

2 citations

Proceedings ArticleDOI
01 Jul 1995
TL;DR: A novel application of uncalibrated stereo reconstruction to Roentgen Stereophotogrammetry Analysis (RSA) and new algorithms are described for automatically localising marker points in X-ray images to sub-pixel accuracy, and using them to reconstruct accurate 3D positions using robust statistical methods.
Abstract: We describe a novel application of uncalibrated stereo reconstruction to Roentgen Stereophotogrammetry Analysis (RSA). In RSA, stereo X-ray images are taken of a bone containing a prosthesis (e.g. a replacement knee) and a number of metal markers. The aim is to recover the relative position of the prosthesis and markers in 3D. Accuracy in previous RSA methods has been limited by two factors: manual feature selection and an assumption that camera calibration parameters are known to high precision - this is not the case in practice. Furthermore, the manual processing is slow and tedious. We report progress towards developing a fully automatic RSA system. New algorithms are described for automatically localising marker points in X-ray images to sub-pixel accuracy, and using them to reconstruct accurate 3D positions using robust statistical methods. Preliminary experiments give excellent results.

2 citations

Proceedings ArticleDOI
04 Feb 1999
TL;DR: The aim of this work is the removal of distracting background patterns from forensic evidence so that the evidence is rendered more visible.
Abstract: The aim of this work is the removal of distracting background patterns from forensic evidence so that the evidenceis rendered more visible. An example is the image of a finger print on a non-periodic background. The methodinvolves registering the image with a control image of the background pattern that we seek to remove. A statisticalcomparison of the registered images identifies the latent mark. The registration of the images involves both a geometric and a photometric component. The geometric registrationis invariant to perspective distortion and the photometric registration invariant to affine colour-space transformations. The algorithm is based on a robust Maximum Likelihood Estimator. Both the registration and the comparisonalgorithms are automatic. The paper briefly explains these algorithms. The method applies in situations where periodic background removal (e.g. Fourier techniques) would not be successful. The method has proven effective in extracting latent fingermark detail overlaying non-periodic backgroundsand examples are shown of its success in removing such backgrounds that would otherwise hamper fingermarkidentification. Indeed, the process has succeeded in rendering visible fingerprints that were totally camouflaged bythe background pattern on bank notes. It is also applicable in removing backgrounds in other forensic cases such asfootprints, for example.Keywords: Forensic images, fingerprints, latent marks, image registration, photometric registration, spatial defor-mations

2 citations

Posted Content
TL;DR: This article proposed a method for visual text recognition without using any paired supervisory data, which enables fully automated, and unsupervised learning from just line-level text-images, and unpaired text-string samples.
Abstract: This work presents a method for visual text recognition without using any paired supervisory data We formulate the text recognition task as one of aligning the conditional distribution of strings predicted from given text images, with lexically valid strings sampled from target corpora This enables fully automated, and unsupervised learning from just line-level text-images, and unpaired text-string samples, obviating the need for large aligned datasets We present detailed analysis for various aspects of the proposed method, namely - (1) impact of the length of training sequences on convergence, (2) relation between character frequencies and the order in which they are learnt, (3) generalisation ability of our recognition network to inputs of arbitrary lengths, and (4) impact of varying the text corpus on recognition accuracy Finally, we demonstrate excellent text recognition accuracy on both synthetically generated text images, and scanned images of real printed books, using no labelled training examples

2 citations

Proceedings Article
06 May 2021
TL;DR: In this article, a Transformer architecture is proposed to temporally align asynchronous subtitles in sign language videos, which is trained on manually annotated alignments covering over 15,000 subtitles that span 17.7 hours of video.
Abstract: The goal of this work is to temporally align asynchronous subtitles in sign language videos. In particular, we focus on sign-language interpreted TV broadcast data comprising (i) a video of continuous signing, and (ii) subtitles corresponding to the audio content. Previous work exploiting such weakly-aligned data only considered finding keyword-sign correspondences, whereas we aim to localise a complete subtitle text in continuous signing. We propose a Transformer architecture tailored for this task, which we train on manually annotated alignments covering over 15K subtitles that span 17.7 hours of video. We use BERT subtitle embeddings and CNN video representations learned for sign recognition to encode the two signals, which interact through a series of attention layers. Our model outputs frame-level predictions, i.e., for each video frame, whether it belongs to the queried subtitle or not. Through extensive evaluations, we show substantial improvements over existing alignment baselines that do not make use of subtitle text embeddings for learning. Our automatic alignment model opens up possibilities for advancing machine translation of sign languages via providing continuously synchronized video-text data.

2 citations


Cited by
More filters
Proceedings ArticleDOI
27 Jun 2016
TL;DR: In this article, the authors proposed a residual learning framework to ease the training of networks that are substantially deeper than those used previously, which won the 1st place on the ILSVRC 2015 classification task.
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.

123,388 citations

Proceedings Article
04 Sep 2014
TL;DR: This work investigates the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting using an architecture with very small 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.
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.

55,235 citations

Proceedings Article
01 Jan 2015
TL;DR: In this paper, the authors investigated the effect of the convolutional network depth on its accuracy in the large-scale image recognition setting and showed that a significant improvement on the prior-art configurations can be achieved by pushing the depth to 16-19 layers.
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.

49,914 citations

Proceedings ArticleDOI
Jia Deng1, Wei Dong1, Richard Socher1, Li-Jia Li1, Kai Li1, Li Fei-Fei1 
20 Jun 2009
TL;DR: A new database called “ImageNet” is introduced, a large-scale ontology of images built upon the backbone of the WordNet structure, much larger in scale and diversity and much more accurate than the current image datasets.
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.

49,639 citations

Book ChapterDOI
05 Oct 2015
TL;DR: Neber et al. as discussed by the authors proposed a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently, which 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.
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 .

49,590 citations