This paper proposes a lightweight yet effective self-supervised depth completion network trained on monocular videos and sparse raw LiDAR measurements only. Specifically, we utilize a multi-stage network architecture, which depends on cheap CNN layers. We introduce a novel guided sparse convolution operator combining sparse and dense data to extract depth features. To mitigate the impact of outliers commonly present in the sparse raw LiDAR data, we adopt a distance-dependent outlier mask that incorporates an elastic threshold mechanism to selectively discard such points. Our experimental results on the KITTI dataset show the favorable trade-off between accuracy and efficiency achieved by our model, reaching state-of-the-art performance on self-supervised depth estimation from few-beams LiDAR (4-beams), depth completion (64-beams) and a few hundred depth points, using a fraction of the parameters. Our code will be available on https://github.com/franky-ciomp/GSCNN/.

Rizhao Fan, F.T. (2023). Lightweight Self-Supervised Depth Estimation with few-beams LiDAR Data. British Machine Vision Association’s (BMVA).

Lightweight Self-Supervised Depth Estimation with few-beams LiDAR Data

Rizhao Fan;Fabio Tosi;Matteo Poggi;Stefano Mattoccia
2023

Abstract

This paper proposes a lightweight yet effective self-supervised depth completion network trained on monocular videos and sparse raw LiDAR measurements only. Specifically, we utilize a multi-stage network architecture, which depends on cheap CNN layers. We introduce a novel guided sparse convolution operator combining sparse and dense data to extract depth features. To mitigate the impact of outliers commonly present in the sparse raw LiDAR data, we adopt a distance-dependent outlier mask that incorporates an elastic threshold mechanism to selectively discard such points. Our experimental results on the KITTI dataset show the favorable trade-off between accuracy and efficiency achieved by our model, reaching state-of-the-art performance on self-supervised depth estimation from few-beams LiDAR (4-beams), depth completion (64-beams) and a few hundred depth points, using a fraction of the parameters. Our code will be available on https://github.com/franky-ciomp/GSCNN/.
2023
The 34th British Machine Vision Conference Proceedings
1
15
Rizhao Fan, F.T. (2023). Lightweight Self-Supervised Depth Estimation with few-beams LiDAR Data. British Machine Vision Association’s (BMVA).
Rizhao Fan, Fabio Tosi, Matteo Poggi, Stefano Mattoccia
File in questo prodotto:
File Dimensione Formato  
0356.pdf

accesso aperto

Tipo: Versione (PDF) editoriale
Licenza: Licenza per accesso libero gratuito
Dimensione 3.08 MB
Formato Adobe PDF
3.08 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11585/962061
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact