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.

Gradient-based learning applied to document recognition

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.

/pdf/going-deeper-with-convolutions-1yobw2o2ds.pdf

Going deeper with convolutions

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 …

I and i

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.

/pdf/histograms-of-oriented-gradients-for-human-detection-3nteinziuq.pdf

Histograms of oriented gradients for human detection

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.

/pdf/rich-feature-hierarchies-for-accurate-object-detection-and-22fvjlggyg.pdf

Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation

Stereo Matching between Three Images by Iterative Refinement in PVS

This paper presents a new multi-lateration algorithm developed by the authors for multiplexed ultrasonic emitters. Our ultrasonic sensor system had a problem that a sampling rate for each emitter had to be decreased to avoid crosstalk as the number of emitter increased. We analyzed the phenomena geometrically and mathematically, and developed a new multi-lateration algorithm that enables our ultrasonic sensor system locate multiple emitters at a time by activating them simultaneously. Simulation results supported the effectiveness of our proposed algorithm and the algorithm implemented in our test system worked perfectly for the experiments conducted in a real environment.

Multi-lateration for multiplexed ultrasonic sensors

An image processing system detects changes in objects, such as damage to automobiles, by comparing a base object model, which depicts the object in an expected condition, to one or more target images of the object in the changed condition The image processing system first processes a target object image to detect one or more predefined landmarks in the target object image and corrects for camera and positional distortions by determining a camera model for the target object image based on the detected landmarks The image processing system then determines contours of the target object within the target object image by overlaying an aligned base object model with the target object image, removes background pixels or other extraneous information based on this comparison, and then performs a statistical processing routine on the identified target object image to detect changes, the likelihood of changes, and/or a quantification of an amount or type of change, to the target object as depicted in the target object image as compared to the base object model

Image processing system to detect changes to target objects using base object models

This paper introduces a new approach to improve the appearance of a prosthetic eye using computer vision and graphics technology. The proposed method can synthesize realistic appearance and movement of a prosthetic eye.

/pdf/appearance-based-prosthetic-eye-4u0u6s1mo5.pdf

Appearance based prosthetic eye

It is important for a robot to be able to self localize during real world operation. This paper presents a verification for a localization method based on stereo cameras and a know grid based map. Verification is achieved by moving the robot through the real world while collecting vision and odometry data over a 200 [m] path. After moving the robot, we manually measured the coordinates of the start point and several way points. The pathway was estimated by the localization process and compared to the measured value to examine the localization effectiveness.

Takeo Kanade

Papers

Stereo Matching between Three Images by Iterative Refinement in PVS

Multi-lateration for multiplexed ultrasonic sensors

Image processing system to detect changes to target objects using base object models

Appearance based prosthetic eye

Verification of stereo vision based localization system