Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

DELTA : Dense Depth from Events and LiDAR Using Transformer's Attention

Brebion, Vincent LU orcid ; Moreau, Julien and Davoine, Franck (2025) 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025 In IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops p.4898-4907
Abstract

Event cameras and LiDARs provide complementary yet distinct data: respectively, asynchronous detections of changes in lighting versus sparse but accurate depth information at a fixed rate. To this day, few works have explored the combination of these two modalities. In this article, we propose a novel neural-network-based method for fusing event and LiDAR data in order to estimate dense depth maps. Our architecture, DELTA, exploits the concepts of self- and cross-attention to model the spatial and temporal relations within and between the event and LiDAR data. Following a thorough evaluation, we demonstrate that DELTA sets a new state of the art in the event-based depth estimation problem, and that it is able to reduce the errors up to... (More)

Event cameras and LiDARs provide complementary yet distinct data: respectively, asynchronous detections of changes in lighting versus sparse but accurate depth information at a fixed rate. To this day, few works have explored the combination of these two modalities. In this article, we propose a novel neural-network-based method for fusing event and LiDAR data in order to estimate dense depth maps. Our architecture, DELTA, exploits the concepts of self- and cross-attention to model the spatial and temporal relations within and between the event and LiDAR data. Following a thorough evaluation, we demonstrate that DELTA sets a new state of the art in the event-based depth estimation problem, and that it is able to reduce the errors up to four times for close ranges compared to the previous SOTA.

(Less)
Please use this url to cite or link to this publication:
author
; and
organization
publishing date
type
Chapter in Book/Report/Conference proceeding
publication status
published
subject
host publication
Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025
series title
IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
pages
10 pages
publisher
IEEE Computer Society
conference name
2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025
conference location
Nashville, United States
conference dates
2025-06-11 - 2025-06-12
external identifiers
  • scopus:105017857550
ISSN
2160-7516
2160-7508
ISBN
9798331599942
DOI
10.1109/CVPRW67362.2025.00482
language
English
LU publication?
yes
id
3c02886f-d1b1-4e8a-9632-41ec060c9a3c
date added to LUP
2025-12-05 12:13:44
date last changed
2025-12-10 17:01:33
@inproceedings{3c02886f-d1b1-4e8a-9632-41ec060c9a3c,
  abstract     = {{<p>Event cameras and LiDARs provide complementary yet distinct data: respectively, asynchronous detections of changes in lighting versus sparse but accurate depth information at a fixed rate. To this day, few works have explored the combination of these two modalities. In this article, we propose a novel neural-network-based method for fusing event and LiDAR data in order to estimate dense depth maps. Our architecture, DELTA, exploits the concepts of self- and cross-attention to model the spatial and temporal relations within and between the event and LiDAR data. Following a thorough evaluation, we demonstrate that DELTA sets a new state of the art in the event-based depth estimation problem, and that it is able to reduce the errors up to four times for close ranges compared to the previous SOTA.</p>}},
  author       = {{Brebion, Vincent and Moreau, Julien and Davoine, Franck}},
  booktitle    = {{Proceedings - 2025 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2025}},
  isbn         = {{9798331599942}},
  issn         = {{2160-7516}},
  language     = {{eng}},
  pages        = {{4898--4907}},
  publisher    = {{IEEE Computer Society}},
  series       = {{IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops}},
  title        = {{DELTA : Dense Depth from Events and LiDAR Using Transformer's Attention}},
  url          = {{http://dx.doi.org/10.1109/CVPRW67362.2025.00482}},
  doi          = {{10.1109/CVPRW67362.2025.00482}},
  year         = {{2025}},
}