Search

Your search keyword '"Li, Shengjia"' showing total 215 results
Start Over Author "Li, Shengjia"
215 results on '"Li, Shengjia"'

2. Reinforced Pedestrian Attribute Recognition with Group Optimization Reward

Author

Ji, Zhong, Hu, Zhenfei, Wang, Yaodong, and Li, Shengjia

Subjects
Computer Science - Computer Vision and Pattern Recognition
Abstract

Pedestrian Attribute Recognition (PAR) is a challenging task in intelligent video surveillance. Two key challenges in PAR include complex alignment relations between images and attributes, and imbalanced data distribution. Existing approaches usually formulate PAR as a recognition task. Different from them, this paper addresses it as a decision-making task via a reinforcement learning framework. Specifically, PAR is formulated as a Markov decision process (MDP) by designing ingenious states, action space, reward function and state transition. To alleviate the inter-attribute imbalance problem, we apply an Attribute Grouping Strategy (AGS) by dividing all attributes into subgroups according to their region and category information. Then we employ an agent to recognize each group of attributes, which is trained with Deep Q-learning algorithm. We also propose a Group Optimization Reward (GOR) function to alleviate the intra-attribute imbalance problem. Experimental results on the three benchmark datasets of PETA, RAP and PA100K illustrate the effectiveness and competitiveness of the proposed approach and demonstrate that the application of reinforcement learning to PAR is a valuable research direction.

Published
2022

3. Analysis of Meshing Contact Characteristics of the Gear Transmission System Based on Data Mining Technology

Author

Li Shengjia, Ma Yali, Zhao Yongsheng, Pu Dajun, and Yan Shidang

Subjects
Gear transmission system, Meshing contact characteristics, Maximum information coefficient, Data mining, Prediction model, Mechanical engineering and machinery, TJ1-1570
Abstract

Aiming at the difficulty of analyzing the meshing contact characteristics of gear transmission system under the big data formed by multiple working conditions and uncertain parameters, a meshing contact characteristics analysis method of gear transmission system based on data mining technology is proposed. Based on the principle of multi-dimensional Gaussian distribution and the finite element model of gear transmission system, the meshing contact characteristic data set of the system is constructed. The correlation between system parameters and meshing contact characteristics is analyzed using the maximum information coefficient, which provides a candidate feature subset for the prediction model. Then, the prediction model of meshing contact characteristics of the system is established using support vector machine and random forest algorithm, which realizes the efficient prediction of meshing contact characteristics of the system. The results show that the prediction error of the prediction model based on support vector machine is the smallest, and the average absolute percentage error is 3.87%, which is far less than the theoretical calculation error. Under the optimal feature subset, the prediction error index of contact characteristics of the prediction model based on support vector machine decreases significantly, and its average absolute percentage error reaches 3.03%, which is 21.71% lower than the prediction error of contact characteristics before optimization. These have verified the accuracy and effectiveness of the proposed method.

Published
2023
Full Text
View/download PDF

22. Semicomplete Multipartite Digraphs Whose Every Arc Is Contained in a Hamiltonian Path

Author

Meng, Wei, Li, Shengjia, Hutchison, David, editor, Kanade, Takeo, editor, Kittler, Josef, editor, Kleinberg, Jon M., editor, Mattern, Friedemann, editor, Mitchell, John C., editor, Naor, Moni, editor, Nierstrasz, Oscar, editor, Rangan, C. Pandu, editor, Steffen, Bernhard, editor, Sudan, Madhu, editor, Terzopoulos, Demetri, editor, Tygar, Doug, editor, Vardi, Moshe Y., editor, Weikum, Gerhard, editor, Shi, Yong, editor, van Albada, Geert Dick, editor, Dongarra, Jack, editor, and Sloot, Peter M. A., editor

Published
2007
Full Text
View/download PDF

Catalog

Books, media, physical & digital resources
See catalog results