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.

Automated construction of robotic manipulation programs

Abstract: Society is becoming more automated with robots beginning to perform most tasks in factories and starting to help out in home and office environments. One of the most important functions of robots is the ability to manipulate objects in their environment. Because the space of possible robot designs, sensor modalities, and target tasks is huge, researchers end up having to manually create many models, databases, and programs for their specific task, an effort that is repeated whenever the task changes. Given a specification for a robot and a task, the presented framework automatically constructs the necessary databases and programs required for the robot to reliably execute manipulation tasks. It includes contributions in three major components that are critical for manipulation tasks. The first is a geometric-based planning system that analyzes all necessary modalities of manipulation planning and offers efficient algorithms to formulate and solve them. This allows identification of the necessary information needed from the task and robot specifications. Using this set of analyses, we build a planning knowledge-base that allows informative geometric reasoning about the structure of the scene and the robot's goals. We show how to efficiently generate and query the information for planners. The second is a set of efficient algorithms considering the visibility of objects in cameras when choosing manipulation goals. We show results with several robot platforms using grippers cameras to boost accuracy of the detected objects and to reliably complete the tasks. Furthermore, we use the presented planning and visibility infrastructure to develop a completely automated extrinsic camera calibration method and a method for detecting insufficient calibration data. The third is a vision-centric database that can analyze a rigid object's surface for stable and discriminable features to be used in pose extraction programs. Furthermore, we show work towards a new voting-based object pose extraction algorithm that does not rely on 2D/3D feature correspondences and thus reduces the early-commitment problem plaguing the generality of traditional vision-based pose extraction algorithms. In order to reinforce our theoric contributions with a solid implementation basis, we discuss the open-source planning environment OpenRAVE, which began and evolved as a result of the work done in this thesis. We present an analysis of its architecture and provide insight for successful robotics software environments.

Three-dimensional scene flow

Abstract: Just as optical flow is the two-dimensional motion of points in an image, scene flow is the three-dimensional motion of points in the world. The fundamental difficulty with optical flow is that only the normal flow can be computed directly from the image measurements, without some form of smoothing or regularization. In this paper, we begin by showing that the same fundamental limitation applies to scene flow; however, many cameras are used to image the scene. There are then two choices when computing scene flow: 1) perform the regularization in the images or 2) perform the regularization on the surface of the object in the scene. In this paper, we choose to compute scene flow using regularization in the images. We describe three algorithms, the first two for computing scene flow from optical flows and the third for constraining scene structure from the inconsistencies in multiple optical flows.

Visual tracking of high DOF articulated structures: an application to human hand tracking

Abstract: Passive sensing of human hand and limb motion is important for a wide range of applications from human-computer interaction to athletic performance measurement High degree of freedom articulated mechanisms like the human hand are difficult to track because of their large state space and complex image appearance This article describes a model-based hand tracking system, called DigitEyes, that can recover the state of a 27 DOF hand model from ordinary gray scale images at speeds of up to 10 Hz

Computer-assisted surgery planner and intra-operative guidance system

Abstract: An apparatus for facilitating the implantation of an artificial component in one of a hip joint, a knee joint, a hand and wrist joint, an elbow joint, a shoulder joint, and a foot and ankle joint. The apparatus includes a pre-operative geometric planner and a pre-operative kinematic biomechanical simulator in communication with the pre-operative geometric planner.

Gradient-based learning applied to document recognition

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.

Going deeper with convolutions

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.

I and i

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 …

Histograms of oriented gradients for human detection

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.

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

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.