VOLDOR: Visual Odometry from Log-logistic Dense Optical flow Residuals

TL;DR

VOLDOR introduces a probabilistic dense visual odometry approach that leverages externally estimated optical flow fields and a log-logistic residual model, achieving high accuracy and modularity on standard benchmarks.

Contribution

The paper presents a novel probabilistic framework for visual odometry using log-logistic residuals, enhancing robustness and modularity over traditional Gaussian-based methods.

Findings

Achieved top results on TUM RGB-D and KITTI benchmarks.

Demonstrated robustness across different optical flow estimators.

Implemented an GPU-efficient, scalable solution.

Abstract

We propose a dense indirect visual odometry method taking as input externally estimated optical flow fields instead of hand-crafted feature correspondences. We define our problem as a probabilistic model and develop a generalized-EM formulation for the joint inference of camera motion, pixel depth, and motion-track confidence. Contrary to traditional methods assuming Gaussian-distributed observation errors, we supervise our inference framework under an (empirically validated) adaptive log-logistic distribution model. Moreover, the log-logistic residual model generalizes well to different state-of-the-art optical flow methods, making our approach modular and agnostic to the choice of optical flow estimators. Our method achieved top-ranking results on both TUM RGB-D and KITTI odometry benchmarks. Our open-sourced implementation is inherently GPU-friendly with only linear computational and storage growth.