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

Video Structure and Terminology.- Multimodal Video Characterization.- Video Summarization.- Visualization Techniques.- Evaluation.- Conclusions.

Multimodal Video Characterization and Summarization

This paper presents the principle, structure and performance of a newly developed MR-compatible force sensor. It employs a new optical micrometry that enables highly accurate and highly sensitive displacement measurement. The sensor accuracy is better than 1.0 %, and the maximum displacement of the detector is about 10 /spl mu/m for a range of the applied force from 0 to 6 N.

Development of an MR-compatible optical force sensor

Picture processing system using a computer complex

This paper presents a method of image-based 3D modeling for intricately-shaped objects, such as a fur, tree leaves and human hair. We formulate the imaging process of these small geometric structures as volume rendering followed by image matting, and prove that the inverse problem can be solved by reducing the nonlinear equations to a large linear system. This estimation, which we call inverse volume rendering, can be performed efficiently through expectation maximization method, even when the linear system is under-constrained owing to data sparseness. We reconstruct object shape by a set of coarse voxels that can model the spatial occupancy inside each voxel. Experimental results show that intricately-shaped objects can successfully be modeled by our proposed method, and the original and other novel view-images of the objects can be synthesized by forward volume rendering.

/pdf/inverse-volume-rendering-approach-to-3d-reconstruction-from-3jb6zni5s0.pdf

Inverse volume rendering approach to 3d reconstruction from multiple images

Meetings are an integral part of business life for any organization. In previous work, we have developed a physical awareness system called CAMEO (Camera Assisted Meeting Event Observer) to record and process the audio/visual information of a meeting. An important task in meeting under-standing is to know who and how many people are attending the meeting. In this paper, we present an automatic approach to detect, track, and cluster people's faces in long video sequences. This is a challenging problem due to the appearance variability of people's faces (illumination, expression, pose, ...). Two main novelties are presented: ? A robust real-time adaptive subspace face tracker which combines color and appearance. ? A temporal subspace clustering algorithm. The effectiveness and robustness of the proposed system is demonstrated over a data set of long videos (i.e. 1 hour).

/pdf/automatic-clustering-of-faces-in-meetings-8m84omel61.pdf

Takeo Kanade

Papers

Multimodal Video Characterization and Summarization

Development of an MR-compatible optical force sensor

Picture processing system using a computer complex

Inverse volume rendering approach to 3d reconstruction from multiple images

Automatic Clustering of Faces in Meetings