CVLab COMPUTER VISION LAB

National Tsing Hua University


Pedestrian Detection

21_cover
Wed, 24 Nov 2010 - wowowo
Video Surveillance from Pedestrian Detection
張幼臻, 王芳瑜, 李郁慈, 蘇德峰, 江振國
Pedestrian detection is helpful to video surveillance when the crime happens. To search a pedestrian from a recorded video database, the background model is built based on a mixture Gaussian model. This model is updated by successive frames to accommodate the lighting change. Then, moving objects are detected via background subtraction. Two classifiers, pixelwise classifier and region-based classifier, are used to remove noises and shadows of objects. After obtaining complete foreground objects, different features, like color, texture, are extracted from each body segment, head, torso and legs. Then, cross matching is performed to ensure effective feature matching. In addition, color consistency is also considered for extremely dark or bright clips.

Car Detection

39_cover
Mon, 04 Feb 2013 - tfsu
EFFICIENT VEHICLE DETECTION WITH ADAPTIVE SCAN BASED ON PERSPECTIVE GEOMETRY
Yu-Chun Chen, Te-Feng Su, and Shang-Hong Lai
Vehicle detection is an important research problem for Advanced Driver Assistance Systems to improve driving safety. Most existing methods are based on the sliding window search framework to locate vehicles in an image. However, such methods usually produce large numbers of false positives and are computationally intensive. In this paper, we propose an efficient vehicle detection algorithm that dramatically reduces the search space based on the perspective geometry of the road. In the training phase, we search a few images to locate all possible vehicle regions by using the standard HOG-based vehicle detector. Pairs of vehicle candidates that satisfy the projective geometry constraints are used to estimate the linear vehicle width model with respect to y coordinates in the image. Then an adaptive scan strategy based on the estimated vehicle width model is proposed to efficiently detect vehicles in an image. Experimental results show that the proposed algorithm provides improved performance in terms of both speed and accuracy compared to standard sliding-windows search strategy.

Image/Video Retargeting

19_cover
Tue, 23 Nov 2010 - Savan
Image Compressibility Assessment and the Application of Structure-Preservng Image Retargeting
Shu-Fan Wang and Shang-Hong Lai
A number of algorithms have been proposed for intelligent image/video retargeting with important content retained as much as possible. In some cases, we can notice that they suffer from artifacts in the resized results, such as ridge or structure twist. In this paper, we suggest that the compressibility of an image should be estimated properly first by analyzing the image structure to determine the optimal scaling factors for the resizing algorithm. To cope with this problem, we propose a compressibility assessment scheme by combining the entropies of image gradient magnitude and orientation distributions. In order to further improve the result, we also present a structure-preserving media retargeting technique that preserves the content and image structure as best as possible. Since we focus on protecting the content structure, a block structure energy is introduced with a top-down strategy to constrain the image structure inside to scale uniformly in either x or y direction. Our experiments demonstrate that the proposed compressibility assessment scheme provides better preservation of content and structure in the resized images/videos than those by the previous methods.
18_cover
Tue, 23 Nov 2010 - wowowo
Fast JND-Based Video Carving with GPU Acceleration for Real-Time Video Retargeting
Chen-Kuo Chiang, Shu-Fan Wang, Yi-Ling Chen and Shang-Hong Lai
We present a fast algorithm for real-time contentaware video retargeting based on the improved seam carving method. The proposed algorithm is designed to be highly parallelizable and suitable for running on a multi-core architecture. First, two novel operators, i.e. seam update and seam split, are introduced to analyze an image for detecting the local and global seams with minimum costs very efficiently. With these operators, parallel processing can be achieved to determine multiple seams simultaneously. In addition, the saliency measure is extended with a Just-Noticeable-Distortion (JND) model which makes the resized video more consistent with human perception. We demonstrate the efficiency of the above new components with a GPU implementation. In addition, the proposed fast seam carving algorithm is extended for video retargeting. To the best of our knowledge, this is the first paper based on the seam carving method to achieve realtime video retargeting on GPU. Experimental results on video sequences of various characteristics are demonstrated to show the superior performance of the proposed algorithm in comparison with the existing content-adaptive image/video resizing methods.
IEEE Transactions on Circuits and Systems for Video Technology, vol. 19, no. 11, pp 1588-1597, Nov. 2009 (pdf)

Computer Vision Applications on Embedded Multi-core Systems

27_cover
Thu, 25 Nov 2010 - tfsu
Parallelized Face Based RMS System on a Multi-core Embedded Computing Platform
蘇德峰、曾聖博、段志學、王書凡、廖家德
The amount of data generated in the world continues to grow at an incredibly high speed. A new framework for the next generation, called the Recognition, Mining and Synthesis (RMS) system, was proposed to make meaningful use of the enormous amount of information on a multi-core processing architecture. Based on the same concept, we propose a face RMS system, which consists of face detection, facial expression recognition, and facial expression exaggeration components, for generating exaggerated views of different expressions on an input face image. In this paper, the parallel algorithms of the face RMS system were developed to reduce the execution time on a multi-core embedded system. The experimental results show the robustness of face detection with different scales and expressions in complex environments and an efficient non-linear method for expression exaggeration. The quantitative comparisons indicate the proposed parallelized face RMS system has a significant increase in speedup compared to the single processor implementation on the multi-core embedded platform, which consists of an ARM processor and two DSP cores.

Over-segmentation Based Background Modelling and Foreground Detection

28_cover
Thu, 25 Nov 2010 - tfsu
Over-Segmentation Based Background Modeling and Foreground Detection with Shadow Removal by Using Hierarchical MRFs
Te-Feng Su,Yi-Ling Chen and Shang-Hong Lai
In this paper, we propose a novel over-segmentation based method for the detection of foreground objects from a surveillance video by integrating techniques of background modeling and Markov Random Fields classification. Firstly, we introduce a fast affinity propagation clustering algorithm to produce the over-segmentation of a reference image by taking into account color difference and spatial relationship between pixels. A background model is learned by using Gaussian Mixture Models with color features of the segments to represent the time-varying background scene. Next, each segment is treated as a node in a Markov Random Field and assigned a state of foreground, shadow and background, which is determined by using hierarchical belief propagation. The relationship between neighboring regions is also considered to ensure spatial coherence of segments. Finally, we demonstrate experimental results on several image sequences to show the effectiveness and robustness of the proposed method.
In Proceedings of Asian Conference on Computer Vision (ACCV'10), Nov. 2010. (pdf)