GroCo: Ground Constraint for Metric Self-Supervised Monocular Depth - Archive ouverte HAL
Communication Dans Un Congrès Année : 2024

GroCo: Ground Constraint for Metric Self-Supervised Monocular Depth

Résumé

Monocular depth estimation has greatly improved in the recent years but models predicting metric depth still struggle to generalize across diverse camera poses and datasets. While recent supervised methods mitigate this issue by leveraging ground prior information at inference, their adaptability to self-supervised settings is limited due to the additional challenge of scale recovery. Addressing this gap, we propose in this paper a novel constraint on ground areas designed specifically for the self-supervised paradigm. This mechanism not only allows to accurately recover the scale but also ensures coherence between the depth prediction and the ground prior. Experimental results show that our method surpasses existing scale recovery techniques on the KITTI benchmark and significantly enhances model generalization capabilities. This improvement can be observed by its more robust performance across diverse camera rotations and its adaptability in zero-shot conditions with previously unseen driving datasets such as DDAD.

Fichier principal
Vignette du fichier
ECCV_2024_Paper-8.pdf (6.16 Mo) Télécharger le fichier
Origine Fichiers produits par l'(les) auteur(s)
Licence

Dates et versions

hal-04704025 , version 1 (20-09-2024)

Licence

Identifiants

  • HAL Id : hal-04704025 , version 1

Citer

Aurélien Cecille, Stefan Duffner, Franck Davoine, Thibault Neveu, Rémi Agier. GroCo: Ground Constraint for Metric Self-Supervised Monocular Depth. European Conference on Computer Vision (ECCV), Sep 2024, Milano, Italy. ⟨hal-04704025⟩
0 Consultations
0 Téléchargements

Partager

More