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.

Anomaly detection through registration

Abstract: We study an application of image registration in the medical domain. Based on a 3-D hierarchical deformable registration algorithm, we have developed a prototype system which automatically aligns a standard atlas to a subject's data to create a customized atlas. Combined with domain knowledge, the registration algorithm can also detect asymmetries and abnormal variations in the subject's data that indicate the existence and location of pathologies. We have conducted tests on 106 MRI scans of normal brains, 3 MRI and 1 CT scan of brains with pathologies, with results qualitatively comparable to manual segmentation.

A perception system for a planetary explorer

Abstract: To perform planetary exploration without human supervision, a completely autonomous robot must be able to model its environment and to locate itself while exploring its surroundings. For that purpose, the authors propose a modular perception system for an autonomous explorer. The perception system maintains a consistent internal representation of the observed terrain from multiple sensor views. The representation can be accessed from other modules through queries. The perception system is intended to be used by the Ambler, a six-legged vehicle being built at the authors' university. A partial implementation of the system using a range scanner is presented as well as experimental results on a testbed that includes the sensor, one computer-controlled leg, and obstacles on a sandy surface. >

Virtualized Reality: Being Mobile in a Visual Scene

Abstract: The visual medium evolved from early paintings to the realistic paintings of the classical era to photographs. The medium of moving imagery started with motion pictures. Television and video recording advanced it to show action “live” or capture and playback later. In all of the above media, the view of the scene is determined at the transcription time, independent of the viewer.

Recognizing Facial Actions by Combining Geometric Features and Regional Appearance Patterns

Abstract: In facial expression analysis, two principle approaches to extract facial features are geometric featurebased methods and appearance-based methods such as Gabor filters. In this paper, we combine these approaches in a feature-based system to recognize Facial Action Coding System (FACS) action units (AUs) in a complex database. The geometric facial features (including mouth, eyes, brows, and cheeks) are extracted using multi-state facial component models. After extraction, these features are represented parametricly. The regional facial appearance patterns are captured using a set of multi-scale and multiorientation Gabor wavelet filters at specific locations. For the upper face, we recognize 8 AUs and neutral expression. The database consists of 606 image sequences from 107 adults of European, African, and Asian ancestry. AUs occur both alone and in combinations. Average recogn ition rate is 87.6% by using geometric facial features alone, 32% by using regional appearance patterns alone, 89.6% by combining both features, and 92.7% after refinement. For the lower face, we recognize 13 AUs and neutral expression. The database consists of 514 image sequences from 180 adults of European, African, and Asian ancestry. AUs occur both alone and in combinations. Average recogn ition rate is 84.7% by using geometric facial features alone, 82% by combining both features, and 87.4% after refinement.

A production system for region analysis

Abstract: Production system architectures are useful for knowledge representation. We have applied the architecture to image analysis in the framework of region growing and developed an outdoor-scene analyzer. In this paper, the following three problems are addressed to make production systems workable in scene analysis. 1) Appropriate size of knowledge represented in a. production rule. 2) Reduction of computation. 3) Control of the analysis toward goal.

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.