Learning generalized visual odometry using position-aware optical flow and geometric bundle adjustment

doi:10.1016/j.patcog.2022.109262

Learning generalized visual odometry using position-aware optical flow and geometric bundle adjustment

2023-04-01

8

OA

AI

DOI：10.1016/j.patcog.2022.109262

原文链接

查看原文

摘要

En 中文

Recent visual odometry (VO) methods incorporating geometric algorithm into deep-learning architecture have shown outstanding performance on the challenging monocular VO task. Despite encouraging results are shown, previous methods ignore the requirement of generalization capability under noisy environ-ment and various scenes. To address this challenging issue, this work first proposes a novel optical flow network (PANet). Compared with previous methods that predict optical flow as a direct regression task, our PANet computes optical flow by predicting it into the discrete position space with optical flow prob-ability volume, and then converting it to optical flow. Next, we improve the bundle adjustment mod-ule to fit the self-supervised training pipeline by introducing multiple sampling, ego-motion initializa-tion, dynamic damping factor adjustment, and Jacobi matrix weighting. In addition, a novel normalized photometric loss function is advanced to improve the depth estimation accuracy. The experiments show that the proposed system not only achieves comparable performance with other state-of-the-art self-supervised learning-based methods on the KITTI dataset, but also significantly improves the generaliza-tion capability compared with geometry-based, learning-based and hybrid VO systems on the noisy KITTI and the challenging outdoor (KAIST) scenes.(c) 2022 Elsevier Ltd. All rights reserved.

Keyword:

Visual odometry

Self-supervise learning

Optical flow

Monocular depth estimation

Joint learning

Generalization capability

Learning generalized visual odometry using position-aware optical flow and geometric bundle adjustment

摘要

AI总结

期刊

机构

引用论文