Attenuation-Aware Weighted Optical Flow with Medium Transmission Map for Learning-based Visual Odometry in Underwater terrain

TL;DR

The paper tackles robust monocular visual odometry for autonomous underwater vehicles by leveraging underwater image formation physics to mitigate optical-flow degradation. It extends the TartanVO framework with an attenuation-aware weighting scheme, using a weighted flow $wF^{t+1}_t = F^{t+1}_t \odot T_{norm}$ where $T_{norm}$ is derived from the estimated inverse transmission $T_{inv}$ via $T_{norm} = \alpha (1/T_{inv}) + 1 - \frac{\max(\alpha(1/T_{inv}))}{\beta}$ and $\sigma = \frac{\max(\alpha(1/T_{inv}))}{\beta}$. Ambient light and transmission are estimated with A-Net and T-Net, trained separately and integrated into a pre-trained VO backbone without fine-tuning. Experiments on real underwater datasets show improved Absolute Trajectory Error (ATE) over baselines, confirming the approach enhances VO robustness in hazy underwater scenes while preserving cross-camera generalization and enabling practical deployment on AUVs.

Abstract

This paper addresses the challenge of improving learning-based monocular visual odometry (VO) in underwater environments by integrating principles of underwater optical imaging to manipulate optical flow estimation. Leveraging the inherent properties of underwater imaging, the novel wflow-TartanVO is introduced, enhancing the accuracy of VO systems for autonomous underwater vehicles (AUVs). The proposed method utilizes a normalized medium transmission map as a weight map to adjust the estimated optical flow for emphasizing regions with lower degradation and suppressing uncertain regions affected by underwater light scattering and absorption. wflow-TartanVO does not require fine-tuning of pre-trained VO models, thus promoting its adaptability to different environments and camera models. Evaluation of different real-world underwater datasets demonstrates the outperformance of wflow-TartanVO over baseline VO methods, as evidenced by the considerably reduced Absolute Trajectory Error (ATE). The implementation code is available at: https://github.com/bachzz/wflow-TartanVO

Figures (3)

• Figure 1: Overview of the TartanVO b6 (black arrows) and the proposed wflow-TartanVO (red arrows) architectures
• Figure 2: Example results on 4 different underwater scenes (top to bottom) in Aqualoc b16.
• Figure 3: Visualization of trajectories on x-y plane (top) and x,y,z axes (bottom) of Aqualoc b16 and RGB SubPipe b17