Skip to main content

Lund University Publications

LUND UNIVERSITY LIBRARIES

Sparse Multiview Open-Vocabulary 3D Detection

Moliner, Olivier LU orcid ; Larsson, Viktor LU orcid and Åström, Kalle LU orcid (2025)
Abstract
The ability to interpret and comprehend a 3D scene is essential for many vision and robotics systems. In numerous applications, this involves 3D object detection, i.e. identifying the location and dimensions of objects belonging to a specific category, typically represented as bounding boxes. This has traditionally been solved by training to detect a fixed set of categories, which limits its use. In this work, we investigate open-vocabulary 3D object detection in the challenging yet practical sparse-view setting, where only a limited number of posed RGB images are available as input. Our approach is training-free, relying on pre-trained, off-the-shelf 2D foundation models instead of employing computationally expensive 3D feature fusion or... (More)
The ability to interpret and comprehend a 3D scene is essential for many vision and robotics systems. In numerous applications, this involves 3D object detection, i.e. identifying the location and dimensions of objects belonging to a specific category, typically represented as bounding boxes. This has traditionally been solved by training to detect a fixed set of categories, which limits its use. In this work, we investigate open-vocabulary 3D object detection in the challenging yet practical sparse-view setting, where only a limited number of posed RGB images are available as input. Our approach is training-free, relying on pre-trained, off-the-shelf 2D foundation models instead of employing computationally expensive 3D feature fusion or requiring 3D-specific learning. By lifting 2D detections and directly optimizing 3D proposals for featuremetric consistency across views, we fully leverage the extensive training data available in 2D compared to 3D. Through standard benchmarks, we demonstrate that this simple pipeline establishes a powerful baseline, performing competitively with state-of-the-art techniques in densely sampled scenarios while significantly outperforming them in the sparse-view setting. (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
2025 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)
pages
10 pages
publisher
IEEE
DOI
10.1109/ICCVW69036.2025.00272
language
English
LU publication?
yes
id
d704ec21-b6a0-4db4-b694-2e4be603b2a7
date added to LUP
2026-04-02 13:57:42
date last changed
2026-04-13 12:17:08
@inproceedings{d704ec21-b6a0-4db4-b694-2e4be603b2a7,
  abstract     = {{The ability to interpret and comprehend a 3D scene is essential for many vision and robotics systems. In numerous applications, this involves 3D object detection, i.e. identifying the location and dimensions of objects belonging to a specific category, typically represented as bounding boxes. This has traditionally been solved by training to detect a fixed set of categories, which limits its use. In this work, we investigate open-vocabulary 3D object detection in the challenging yet practical sparse-view setting, where only a limited number of posed RGB images are available as input. Our approach is training-free, relying on pre-trained, off-the-shelf 2D foundation models instead of employing computationally expensive 3D feature fusion or requiring 3D-specific learning. By lifting 2D detections and directly optimizing 3D proposals for featuremetric consistency across views, we fully leverage the extensive training data available in 2D compared to 3D. Through standard benchmarks, we demonstrate that this simple pipeline establishes a powerful baseline, performing competitively with state-of-the-art techniques in densely sampled scenarios while significantly outperforming them in the sparse-view setting.}},
  author       = {{Moliner, Olivier and Larsson, Viktor and Åström, Kalle}},
  booktitle    = {{2025 IEEE/CVF International Conference on Computer Vision Workshops (ICCVW)}},
  language     = {{eng}},
  publisher    = {{IEEE}},
  title        = {{Sparse Multiview Open-Vocabulary 3D Detection}},
  url          = {{http://dx.doi.org/10.1109/ICCVW69036.2025.00272}},
  doi          = {{10.1109/ICCVW69036.2025.00272}},
  year         = {{2025}},
}