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

抗原变异可使得多种致病微生物易于逃避宿主免疫应答。表达在感染红细胞表面的恶性疟原虫红细胞表面蛋白1（PfPMP1）与感染红细胞、内皮细胞、树突状细胞以及胎盘的单个或多个受体作用，在黏附及免疫逃避中起关键的作用。每个单倍体基因组var基因家族编码约60种成员，通过启动转录不同的var基因变异体为抗原变异提供了分子基础。

疟原虫var基因转换速率变化导致抗原变异[英]／Paul H, Robert P, Christodoulou Z, et al//Proc Natl Acad Sci U S A

Presents a theoretically very simple, yet efficient, multiresolution approach to gray-scale and rotation invariant texture classification based on local binary patterns and nonparametric discrimination of sample and prototype distributions. The method is based on recognizing that certain local binary patterns, termed "uniform," are fundamental properties of local image texture and their occurrence histogram is proven to be a very powerful texture feature. We derive a generalized gray-scale and rotation invariant operator presentation that allows for detecting the "uniform" patterns for any quantization of the angular space and for any spatial resolution and presents a method for combining multiple operators for multiresolution analysis. The proposed approach is very robust in terms of gray-scale variations since the operator is, by definition, invariant against any monotonic transformation of the gray scale. Another advantage is computational simplicity as the operator can be realized with a few operations in a small neighborhood and a lookup table. Experimental results demonstrate that good discrimination can be achieved with the occurrence statistics of simple rotation invariant local binary patterns.

/pdf/multiresolution-gray-scale-and-rotation-invariant-texture-1b1y2fta4g.pdf

Multiresolution gray-scale and rotation invariant texture classification with local binary patterns

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 compare R-CNN to OverFeat, a recently proposed sliding-window detector based on a similar CNN architecture. We find that R-CNN outperforms OverFeat by a large margin on the 200-class ILSVRC2013 detection dataset. Source code for the complete system is available at this http URL.

Rich feature hierarchies for accurate object detection and semantic segmentation

Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

This paper concerns the pose determination and recognition of vehicles in traffic scenes, which under normal conditions stand on the ground-plane. Novel linear and closed-form algorithms are described for pose determination from an arbitrary number of known line matches. A form of the generalised Hough transform is used in conjuction with explicit probability-based voting models to find consistent matches. The algorithms are fast and robust. They cope well with complex outdoor scenes.

Pose determination and recognition of vehicles in traffic scenes

Traditional steganalysis methods usually rely on handcrafted features. However, with the rapid development of advanced steganography, manual design of complex features has become increasingly difficult. In this paper, we propose a new paradigm for steganalysis based on the concept of feature learning. In our method, Convolutional Neural Network (CNN) is used to automatically learn features for steganalysis. To make CNN work better for steganalysis, we incorporate domain knowledge of steganalysis (i.e. enhancing stego noise and exploiting nearby dependencies) when designing the CNN architectures. We further propose to use model combination to boost the performance of CNN based method. Additionally, a cropping strategy is proposed to enable the CNN based model to deal with arbitrary input image sizes. We demonstrate the effectiveness of the proposed method against state-of-the-art spatial domain steganographic algorithms such as HUGO, WOW, S-UNIWARD, MiPOD, and HILL-CMD. To help understand the learned features from CNN, we provide visualizations of the learned filters and feature maps. Finally, we also provide quantitative analysis of the learned features from convolutional layers.

Feature learning for steganalysis using convolutional neural networks

Cast shadow on the background is generated by an object moving between a light source and the background. The position and illumination of the source always change with time, while the background is stable. Therefore, features connected with light source always change with time, such as geometry and color. In this paper, we present a shadow removal method by homomorphic model to extract surface reflectance component, which is only connected with background of the scene and is robust to change of light source. We assume that reflectance component fits Gaussian distribution, and then use GMM to model it. Experimental results show that, except dealing with shadow, our method is not sensitive to the change of illumination.

/pdf/cast-shadow-removal-with-gmm-for-surface-reflectance-4nwlmmvlxy.pdf

Cast Shadow Removal with GMM for Surface Reflectance Component

In this paper, a dynamic model of car motion is proposed in which the turn of the steering wheel and the distance between the front and rear wheel are taken into account. Extended Kalman Filter (EKF) is widely used in visual tracking systems. However, because there is no direct link between the behaviour of the driver who controls the motion of the car and the assumed dynamic model, the traditional EKF does not perform well when the car carries out a complicated manoeuvre. In order to reduce the sensitivity of the filter to the model uncertainty, we use a modified EKF by adding a new objective function. Experimental results show that the performance of our approach is much better than the traditional EKF and its recent variant.

/pdf/visual-vehicle-tracking-using-an-improved-ekf-5bgl2cg6ju.pdf

Visual Vehicle Tracking Using An Improved EKF

The present disclosure relates to a gait recognition method based on deep learning, which comprises recognizing an identity of a person in a video according to the gait thereof through dual-channel convolutional neural networks sharing weights by means of the strong learning capability of the deep learning convolutional neural network. Said method is quite robust to gait changes across a large view, which can effectively solve the problem of low precision in cross-view gait recognition existing with the prior art gait recognition technology. Said method can be widely used in scenarios having video monitors, such as security monitoring in airports and supermarkets, person recognition, criminal detection, etc.

Tieniu Tan

Papers

Pose determination and recognition of vehicles in traffic scenes

Feature learning for steganalysis using convolutional neural networks

Cast Shadow Removal with GMM for Surface Reflectance Component

Visual Vehicle Tracking Using An Improved EKF

Gait recognition method based on deep learning