Pixel-Wise Prediction based Visual Odometry via Uncertainty Estimation
This is an incremental improvement for robotics and autonomous systems, focusing on enhancing visual odometry accuracy through uncertainty handling.
The paper tackles visual odometry by introducing a dense pixel-wise prediction method (PWVO) that uses uncertainty estimation to filter noisy regions, achieving favorable results in experiments.
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 is able to deliver favorable results. In addition, our analyses validate the effectiveness of the designs adopted in PWVO, and demonstrate that the uncertainty maps estimated by PWVO is capable of capturing the noises in its input observations.