Joint direct estimation of 3D geometry and 3D motion using spatio temporal gradients

TL;DR

This paper introduces a direct method for 3D structure and motion estimation from image gradients, avoiding optical flow computation, leading to improved accuracy in synthetic and real-world datasets.

Contribution

It proposes a novel normal flow based approach that reformulates the positive-depth constraint for better 3D motion and structure estimation.

Findings

Outperforms existing normal flow based methods in accuracy

Achieves better results than traditional optical flow based methods in most tests

Provides highly accurate 3D geometry recovery

Abstract

Conventional image motion based structure from motion methods first compute optical flow, then solve for the 3D motion parameters based on the epipolar constraint, and finally recover the 3D geometry of the scene. However, errors in optical flow due to regularization can lead to large errors in 3D motion and structure. This paper investigates whether performance and consistency can be improved by avoiding optical flow estimation in the early stages of the structure from motion pipeline, and it proposes a new direct method based on image gradients (normal flow) only. The main idea lies in a reformulation of the positive-depth constraint, which allows the use of well-known minimization techniques to solve for 3D motion. The 3D motion estimate is then refined and structure estimated adding a regularization based on depth. Experimental comparisons on standard synthetic datasets and the real-world driving benchmark dataset KITTI using three different optic flow algorithms show that the method achieves better accuracy in all but one case. Furthermore, it outperforms existing normal flow based 3D motion estimation techniques. Finally, the recovered 3D geometry is shown to be also very accurate.