CoChat
Sign Up
Cited by 111Advanced Neural Network ApplicationsAdvanced Image and Video Retrieval TechniquesDomain Adaptation and Few-Shot Learning

Omni3D: A Large Benchmark and Model for 3D Object Detection in the Wild

Open access
Brazil, Garrick, Kumar, Abhinav, Straub, Julian, Ravi, Nikhila, Johnson, Justin, Gkioxari, Georgia·2023·2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
DOI 10.1109/cvpr52729.2023.01264
Open this paper in CoChat — free accountOpen PDF

Abstract

Recognizing scenes and objects in 3D from a single image is a longstanding goal of computer vision with applications in robotics and AR/VR. For 2D recognition, large datasets and scalable solutions have led to unprecedented advances. In 3D, existing benchmarks are small in size and approaches specialize in few object categories and specific domains, e.g. urban driving scenes. Motivated by the success of 2D recognition, we revisit the task of 3D object detection by introducing a large benchmark, called Omni3D. Omni3DRE-purposes and combines existing datasets resulting in 234k images annotated with more than 3 million instances and 98 categories. 3D detection at such scale is challenging due to variations in camera intrinsics and the rich diversity of scene and object types. We propose a model, called Cube R-CNN, designed to generalize across camera and scene types with a unified approach. We show that Cube R-CNN outperforms prior works on the larger Omni3D and existing benchmarks. Finally, we prove that Omni3D is a powerful dataset for 3D object recognition and show that it improves single-dataset performance and can accelerate learning on new smaller datasets via pre-training.11We release the Omni3D benchmark and Cube R-CNN models at https://github.com/facebookresearch/omni3d.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1109/cvpr52729.2023.01264

Cites

14 of 14 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 6

  1. [2][2] Adel Ahmadyan, Liangkai Zhang, Artsiom Ablavatski, Jianing Wei, and Matthias Grundmann. Objectron: A large scale dataset of object-centric videos in the wild with pose annotations. CVPR , 2021. 2, 3, 6, 9, 14 -
  2. [3][3] Armen Avetisyan, Manuel Dahnert, Angela Dai, Manolis Savva, Angel X Chang, and Matthias Nießner. Scan2CAD: Learning CAD model alignment in RGB-D scans. In Proceedings of the IEEE/CVF Conference on computer vision and pattern recognition , pages 2614-2623, 2019. 2 -
  3. [5][5] Bradford Barber, David Dobkin, and Hannu Huhdanpaa. The quickhull algorithm for convex hulls. ACM Transactions on Mathematical Software (TOMS) , 1996. 6 -
  4. [6][6] Gilad Baruch, Zhuoyuan Chen, Afshin Dehghan, Tal Dimry, Yuri Feigin, Peter Fu, Thomas Gebauer, Brandon Joffe, Daniel Kurz, Arik Schwartz, and Elad Shulman. ARKitScenes - a [2 https://omni3d.garrickbrazil.com/#demo](https://omni3d.garrickbrazil.com/#demo) Figure 12. Cube R-CNN on OMNI3D test. We show the input image, the 3D predictions overlaid on the image and a top view. We show examples from SUN RGB-D [73], ARKitScenes [6], Hypersim [68], Objectron [2], nuScenes [9], and KITTI [22]. <!-- image --> - diverse real-world dataset for 3D indoor scene understanding using mobile RGB-D data. In NeurIPS Datasets and Benchmarks Track , 2021. 2, 3, 9, 14 -
  5. [7][7] Garrick Brazil and Xiaoming Liu. M3D-RPN: Monocular 3D region proposal network for object detection. In ICCV , 2019. 1, 2, 5, 7, 11, 13 -
  6. [8][8] Garrick Brazil, Gerard Pons-Moll, Xiaoming Liu, and Bernt Schiele. Kinematic 3D object detection in monocular video. In ECCV , 2020. 2 -

Cited by

Sorted by citations

  • Qwen3-VL Technical Report

    Bai, S., Cai, Y., Chen, R., Chen, K., Chen, X., Cheng, Z., Deng, L., Ding, W., Gao, C., Ge, C., et al. (2025) · arXiv · doi:10.48550/arxiv.2511.21631 · Cited by 27 · see citing works

    “To ensure a fair comparison with existing VLMs, we set the IoU threshold to 0.15 and report mAP@0.15 on the Omni3D test set, with detection confidence fixed at 1.0.”
    Influential

Authors
Brazil, G., Kumar, A., Straub, J., Ravi, N., Johnson, J., & Gkioxari, G.
Affiliations
  • Meta (United States) (US)
  • Michigan State University (US)
  • California Institute of Technology (US)
ORCID
Venue
2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Publisher
IEEE
Published
2023-06
Pages
13154-13164
ISSN
1063-6919 (ISSN-L)
DOI
10.1109/cvpr52729.2023.01264
OpenAlex
W4386075853
Cited by
111
References
14
Citations by year
2026: 31 · 2025: 36 · 2024: 35 · 2023: 9
FWCI
11.29
Type
conference paper
Language
en
Open access
Licence
Locations
  • IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings (publishedVersion)
  • arXiv (Cornell University) (submittedVersion)
  • CaltechAUTHORS (California Institute of Technology) (submittedVersion, cc-by-nc-sa)
  • arXiv (Cornell University)
Topics
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
  • Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Image and Video Retrieval Techniques
  • Physical Sciences › Computer Science › Artificial Intelligence › Domain Adaptation and Few-Shot Learning
Keywords
3D object detection, 3D object recognition, large-scale benchmark, pre-training

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review