Composite-mask GAN based on refined optical flow and disparity map for SLAM visual odometry

作     者：JI Yuehui JIANG Jingwei LIU Junjie SONG Yu GAO Qiang 

作者机构：School of Electrical Engineering and Automation， Tianjin University of Technology Tianjin Key Laboratory of Control Theory and Application for Complex Systems 

出 版 物：《Optoelectronics Letters》 (光电子快报(英文))

年 卷 期：2025年

学科分类：08[工学] 080203[工学-机械设计及理论] 0802[工学-机械工程] 

基　　金：supported by the Program of Graduate Education and Teaching Reform in Tianjin University of Technology YBXM2204,ZDXM2202 the National Natural Science Foundation of China under Grant (No.62203331) the National Natural Science Foundation of China under Grant (No.62103299) 

摘      要：Although deep learning methods have been widely applied in slam visual odometry over the past decade with impressive improvements， the accuracy remains limited in complex dynamic environments. In this paper， a composite mask-based generative adversarial network is introduced to predict camera motion and binocular depth maps. Specifically， a perceptual generator is constructed to obtain the corresponding parallax map and optical flow from between two neighboring frames. Then， an iterative pose improvement strategy is proposed to improve the accuracy of pose estimation. Finally， a composite mask is embedded in the discriminator to sense structural deformation in the synthesized virtual image， thereby increasing the overall structural constraints of the network model， improving the accuracy of camera pose estimation， and reducing drift issues in the Visual Odometer. Detailed quantitative and qualitative evaluations on the KITTI dataset show that the proposed framework outperforms existing conventional， supervised learning and unsupervised depth VO methods， providing better results in both pose estimation and depth estimation.