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.

Correlation Filters for Object Alignment

Abstract: Alignment of 3D objects from 2D images is one of the most important and well studied problems in computer vision. A typical object alignment system consists of a landmark appearance model which is used to obtain an initial shape and a shape model which refines this initial shape by correcting the initialization errors. Since errors in landmark initialization from the appearance model propagate through the shape model, it is critical to have a robust landmark appearance model. While there has been much progress in designing sophisticated and robust shape models, there has been relatively less progress in designing robust landmark detection models. In this paper we present an efficient and robust landmark detection model which is designed specifically to minimize localization errors thereby leading to state-of-the-art object alignment performance. We demonstrate the efficacy and speed of the proposed approach on the challenging task of multi-view car alignment.

Geometric camera calibration using systems of linear equations

Abstract: Geometric camera calibration is the process of determining a mapping between points in world coordinates and the corresponding image locations of the points. In previous methods, calibration typically involved the iterative solution to a system of nonlinear equations. A method is presented for performing camera calibration that provides a complete, accurate solution, using only linear systems of equations. By using two calibration planes, a line-of-sight vector is defined for each pixel in the image. The effective focal point of a camera can be obtained by solving the system that defines the intersection point of the line-of-sight vectors. Once the focal point has been determined, a complete camera model can be obtained with a straightforward least-squares procedure. This method of geometric camera calibration has the advantages of being accurate, efficient, and practical for a wide variety of applications. >

A locally adaptive window for signal matching

Abstract: This article presents a signal matching algorithm that can select an appropriate window size adaptively so as to obtain both precise and stable estimation of correspondences. Matching two signals by calculating the sum of squared differences (SSD) over a certain window is a basic technique in computer vision. Given the signals and a window, there are two factors that determine the difficulty of obtaining precise matching. The first is the variation of the signal within the window, which must be large enough, relative to noise, that the SSD values exhibit a clear and sharp minimum at the correct disparity. The second factor is the variation of disparity within the window, which must be small enough that signals of corresponding positions are duly compared. These two factors present conflicting requirements to the size of the matching window, since a larger window tends to increase the signal variation, but at the same time tends to include points of different disparity. A window size must be adaptively selected depending on local variations of signal and disparity in order to compute a most-certain estimate of disparity at each point. There has been little work on a systematic method for automatic window-size selection. The major difficulty is that, while the signal variation is measurable from the input, the disparity variation is not, since disparities are what we wish to calculate. We introduce here a statistical model of disparity variation within a window, and employ it to establish a link between the window size and the uncertainty of the computed disparity. This allows us to choose the window size that minimizes uncertainty in the disparity computed at each point. This article presents a theory for the model and the resultant algorithm, together with analytical and experimental results that demonstrate their effectiveness.

Vision-Based Object Registration for Real-Time Image Overlay

Abstract: This paper presents computer vision based techniques for object registration, real-time tracking, and image overlay. The capability can be used to superimpose registered images such as those from CT or MRI onto a video image of a patient’s body. Real-time object registration enables an image to be overlaid consistently onto objects even while the object or the viewer is moving. The video image of a patient’s body is used as input for object registration. Reliable real-time object registration at frame rate (30 Hz) is realized by a combination of techniques, including template matching based feature detection, feature correspondence by geometric constraints, and pose calculation of objects from feature positions in the image. Two types of image overlay systems are presented. The first one registers objects in the image and projects preoperative model data onto a raw camera image. The other computes the position of image overlay directly from 2D feature positions without any prior models. With the techniques developed in this paper, interactive video, which transmits images of a patient to the expert and sends them back with some image overlay, can be realized.

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.