Pixel-Wise Prediction Based Visual Odometry via Uncertainty Estimation

Hao-Wei Chen, Ting-Hsuan Liao, Hsuan-Kung Yang, Chun-Yi Lee; Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), 2023, pp. 2518-2528

Abstract


This paper introduces pixel-wise prediction based visual odometry(PWVO), which is a dense prediction task that evaluates the values of translation and rotation for every pixel in its input observations. PWVO employs uncertainty estimation to identify the noisy regions in the input observations, and adopts a selection mechanism to integrate pixel-wise predictions based on the estimated uncertainty maps to derive the final translation and rotation. In order to train PWVO in a comprehensive fashion, we further develop a data generation workflow for generating synthetic training data. The experimental results show that PWVO isable to deliver favorable results. In addition, our analyses validate the effectiveness of the designs adopted in PWVO, and demonstrate that the uncertainty mapsestimated by PWVO is capable of capturing the noises in its input observations.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{Chen_2023_WACV, author = {Chen, Hao-Wei and Liao, Ting-Hsuan and Yang, Hsuan-Kung and Lee, Chun-Yi}, title = {Pixel-Wise Prediction Based Visual Odometry via Uncertainty Estimation}, booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)}, month = {January}, year = {2023}, pages = {2518-2528} }