scispace - formally typeset
Search or ask a question
Author

Takeo Kanade

Bio: Takeo Kanade is an academic researcher from Carnegie Mellon University. The author has contributed to research in topics: Motion estimation & Image processing. The author has an hindex of 147, co-authored 799 publications receiving 103237 citations. Previous affiliations of Takeo Kanade include National Institute of Advanced Industrial Science and Technology & Hitachi.


Papers
More filters
Proceedings ArticleDOI
13 Oct 1999
TL;DR: A computational model is developed and verified for the image formation process of differential interference contrast microscopy and it is planned to use this model to reconstruct the properties of unknown specimens.
Abstract: Biologists often use differential interference contrast (DIC) microscopy to study live cells. However, they are limited to qualitative observations due to the inherent nonlinear relation between the object properties and image intensity. As a first step towards quantitatively measuring optical properties of objects from DIC images, we develop and verify a computational model for the image formation process. Next, we plan to use this model to reconstruct the properties of unknown specimens.

1 citations

Journal ArticleDOI
TL;DR: This CVIU special issue gathers very recent and various works n assistive computer vision and robotics that have applications in robotics as multi-modal human-robot interaction, autonomous navigaion, object usage, place recognition, robotic manipulator, egocenric vision.

1 citations

Book ChapterDOI
01 Jan 2003
TL;DR: This work has developed a system that tracks a person in real-time and adjusts the pan, tilt, zoom and focus of each camera to acquire synchronized multi-view video of a person moving through the scene.
Abstract: For applications in human identification, activity recognition, 3D reconstruction, entertainment and sports, it is often desirable to capture a set of synchronized video sequences of a person from multiple camera viewpoints (see Figure 8-1). One way to achieve this is to set up a ring of cameras all statically aimed at a single point in space, and to have an actor perform at this fixation point while the video footage is shot. This is the method used to create spectacular special effects in the movie The Matrix, where playing back frames from a single time step, across all cameras, yields the appearance of freezing the action in time while a virtual camera flies around the scene. However, in surveillance or sports applications it is not possible to predict beforehand the precise location where an interesting activity will occur, and therefore it is necessary to dynamically adjust the fixation point of multiple camera views. We have developed a system that tracks a person in real-time and adjusts the pan, tilt, zoom and focus of each camera to acquire synchronized multi-view video of a person moving through the scene.

1 citations

Journal ArticleDOI
TL;DR: This paper proposes a method that estimate road contour using delineators, which was able to classify the road contours by high accuracy.
Abstract: Drivers of vehicles focus their gaze in the direction of movement, driver guesses an optimum route using the white road line and the delineators. However, the range that can be clearly seen in the headlights is limited, it is difficult to guess the optimum route. This paper proposes a method that estimate road contour using delineators. The road contours are estimated from the 3D positions of delineators located on the sides of roads, which are extracted using a circle detection filter. Then, clothoid curve is applied to the delineators and the parameters of clothoid curve are obtained. This classifies the parameter into four kinds of curves using support vector machine. In simulation experiment, we create a virtual road. A classification rate was 86.9 %. Our method was able to classify the road contour by high accuracy.

1 citations

Proceedings Article
01 Jan 2007
TL;DR: A process for preserving green colored plant tissues and in particular coniferous needles, holly and low fiber leaves such as mosses, lichens and ferns in which selected leaves are immersed in a solution to permanently retain and biologically fix the green color of the leaves.
Abstract: A process for preserving green colored plant tissues and in particular coniferous needles, holly and low fiber leaves such as mosses, lichens and ferns in which selected leaves are immersed in a solution comprising (by volume) 35-45% water, 20-30% 2-propanol, 5-12% propionic acid, 5-10% sulphurous acid, 5-10% formalin, 2.5-5% formic acid, 1-5% ethylene glycol, and optionally minor amounts of compounds selected from the group consisting of cupric sulphate, cupric chloride, 20-20-20 fertilizer, citric acid, DBE, magnesium sulphate, acetic acid, cupric acetate, cupric nitrate, sodium phosphate, sodium sulfite, butylated hydroxytolulene and glycerol, for a sufficient time to exchange the naturally occurring water in the tissues with the "chemical water" of the solution and thereby permanently retain and biologically fix the green color of the leaves.

1 citations


Cited by
More filters
Journal ArticleDOI
01 Jan 1998
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.

42,067 citations

Proceedings ArticleDOI
07 Jun 2015
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
Abstract: We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.

40,257 citations

Journal ArticleDOI

[...]

08 Dec 2001-BMJ
TL;DR: There is, I think, something ethereal about i —the square root of minus one, which seems an odd beast at that time—an intruder hovering on the edge of reality.
Abstract: There is, I think, something ethereal about i —the square root of minus one. I remember first hearing about it at school. It seemed an odd beast at that time—an intruder hovering on the edge of reality. Usually familiarity dulls this sense of the bizarre, but in the case of i it was the reverse: over the years the sense of its surreal nature intensified. It seemed that it was impossible to write mathematics that described the real world in …

33,785 citations

Proceedings ArticleDOI
20 Jun 2005
TL;DR: It is shown experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection, and the influence of each stage of the computation on performance is studied.
Abstract: We study the question of feature sets for robust visual object recognition; adopting linear SVM based human detection as a test case. After reviewing existing edge and gradient based descriptors, we show experimentally that grids of histograms of oriented gradient (HOG) descriptors significantly outperform existing feature sets for human detection. We study the influence of each stage of the computation on performance, concluding that fine-scale gradients, fine orientation binning, relatively coarse spatial binning, and high-quality local contrast normalization in overlapping descriptor blocks are all important for good results. The new approach gives near-perfect separation on the original MIT pedestrian database, so we introduce a more challenging dataset containing over 1800 annotated human images with a large range of pose variations and backgrounds.

31,952 citations

Proceedings ArticleDOI
23 Jun 2014
TL;DR: RCNN as discussed by the authors combines CNNs with bottom-up region proposals to localize and segment objects, and when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost.
Abstract: Object detection performance, as measured on the canonical PASCAL VOC dataset, has plateaued in the last few years. The best-performing methods are complex ensemble systems that typically combine multiple low-level image features with high-level context. In this paper, we propose a simple and scalable detection algorithm that improves mean average precision (mAP) by more than 30% relative to the previous best result on VOC 2012 -- achieving a mAP of 53.3%. Our approach combines two key insights: (1) one can apply high-capacity convolutional neural networks (CNNs) to bottom-up region proposals in order to localize and segment objects and (2) when labeled training data is scarce, supervised pre-training for an auxiliary task, followed by domain-specific fine-tuning, yields a significant performance boost. Since we combine region proposals with CNNs, we call our method R-CNN: Regions with CNN features. We also present experiments that provide insight into what the network learns, revealing a rich hierarchy of image features. Source code for the complete system is available at http://www.cs.berkeley.edu/~rbg/rcnn.

21,729 citations