No abstract

Pedestrian detection is widely applied in surveillance, autonomous robotic navigation, and automotive safety. However, there are many occlusion problems in real life. This paper summarizes the research progress of pedestrian detection technology with occlusion. First, according to different occlusion, it can be divided into two categories: inter-class occlusion and intra-class occlusion. Second, it summarizes the traditional method and deep learning method to deal with occlusion. Furthermore, the main ideas and core problems of each method model are analyzed and discussed. Finally, the paper gives an outlook on the problems to be solved in the future development of pedestrian detection technology with occlusion.

The paper summarizes the research progress on critical region recognition and deep metric learning to achieve accurate clothing image retrieval in cross-domain situations. Critical region recognition is of great value for the clothing feature extraction, effectively improving retrieval accuracy. The accuracy will decrease when solving difficult samples with similar features but different categories. Nowadays, deep metric learning is an effective way to solve this problem, which utilizes the optimization of different loss functions and ensemble network to strengthen the discrimination of clothing features. Therefore, through comparison of the experimental results of different algorithms and analysis of the accuracy of cross-domain clothing retrieval, it is demonstrated that the improvement of the retrieval accuracy in the future mainly depends on clothing important feature extraction and clothing feature discrimination.

Pedestrian detection technology is the use of computer vision technology to determine whether there are pedestrians in static images or video images, and accurately mark pedestrians. Intelligent vehicles, intelligent monitoring, human behaviour analysis and other aspects are related to pedestrian detection technology, so pedestrian detection technology has been widely concerned by all walks of life. However, in some application environments, the network setting is flexible and the location of the device can be changed at any time, so it is difficult to establish a pedestrian detection system by wired way, and the pedestrian detection system can only be established through WSN. It is necessary to consider how to complete pedestrian detection through intelligent terminal under the condition of limited computing power. Many researchers have put forward a lot of algorithms, which are constantly optimized, but there are still many problems to be solved in practical application. In this paper, aiming at the intelligent terminal in WSN environment with limited computing power, a pedestrian detection system based on HOG+SVM is designed, which can detect moving pedestrians in video in real time. The detection system is mainly composed of detection algorithm and video acquisition module. After the test of INRIA and CVC data set, the accuracy of the detection algorithm in this paper is 89.06%. This method is tested under the intelligent terminal with limited computing power, with a good detection effect.

No abstract