English

Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation

Computer Vision and Pattern Recognition 2023-02-03 v1 Machine Learning Robotics

Abstract

Dense depth estimation is essential to scene-understanding for autonomous driving. However, recent self-supervised approaches on monocular videos suffer from scale-inconsistency across long sequences. Utilizing data from the ubiquitously copresent global positioning systems (GPS), we tackle this challenge by proposing a dynamically-weighted GPS-to-Scale (g2s) loss to complement the appearance-based losses. We emphasize that the GPS is needed only during the multimodal training, and not at inference. The relative distance between frames captured through the GPS provides a scale signal that is independent of the camera setup and scene distribution, resulting in richer learned feature representations. Through extensive evaluation on multiple datasets, we demonstrate scale-consistent and -aware depth estimation during inference, improving the performance even when training with low-frequency GPS data.

Keywords

Cite

@article{arxiv.2103.02451,
  title  = {Multimodal Scale Consistency and Awareness for Monocular Self-Supervised Depth Estimation},
  author = {Hemang Chawla and Arnav Varma and Elahe Arani and Bahram Zonooz},
  journal= {arXiv preprint arXiv:2103.02451},
  year   = {2023}
}

Comments

Accepted at 2021 IEEE International Conference on Robotics and Automation (ICRA)

R2 v1 2026-06-23T23:42:50.676Z