Improving Distant 3D Object Detection Using 2D Box Supervision

• URL: http://arxiv.org/abs/2403.09230v1
• Date: Thu, 14 Mar 2024 09:54:31 GMT
• Title: Improving Distant 3D Object Detection Using 2D Box Supervision
• Authors: Zetong Yang, Zhiding Yu, Chris Choy, Renhao Wang, Anima Anandkumar, Jose M. Alvarez,
• Abstract summary: We propose LR3D, a framework that learns to recover the missing depth of distant objects. Our framework is general, and could widely benefit 3D detection methods to a large extent.
• Score: 97.80225758259147
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Improving the detection of distant 3d objects is an important yet challenging task. For camera-based 3D perception, the annotation of 3d bounding relies heavily on LiDAR for accurate depth information. As such, the distance of annotation is often limited due to the sparsity of LiDAR points on distant objects, which hampers the capability of existing detectors for long-range scenarios. We address this challenge by considering only 2D box supervision for distant objects since they are easy to annotate. We propose LR3D, a framework that learns to recover the missing depth of distant objects. LR3D adopts an implicit projection head to learn the generation of mapping between 2D boxes and depth using the 3D supervision on close objects. This mapping allows the depth estimation of distant objects conditioned on their 2D boxes, making long-range 3D detection with 2D supervision feasible. Experiments show that without distant 3D annotations, LR3D allows camera-based methods to detect distant objects (over 200m) with comparable accuracy to full 3D supervision. Our framework is general, and could widely benefit 3D detection methods to a large extent.

Related papers

• General Geometry-aware Weakly Supervised 3D Object Detection [62.26729317523975]
  A unified framework is developed for learning 3D object detectors from RGB images and associated 2D boxes. Experiments on KITTI and SUN-RGBD datasets demonstrate that our method yields surprisingly high-quality 3D bounding boxes with only 2D annotation.
  arXiv  Detail & Related papers  (2024-07-18T17:52:08Z)
• Enhancing 3D Object Detection with 2D Detection-Guided Query Anchors [6.3557174349423455]
  We present a novel query generating approach termed QAF2D, which infers 3D query anchors from 2D detection results. The largest improvement that QAF2D can bring about on the nuScenes validation subset is $2.3%$ NDS and $2.7%$ mAP.
  arXiv  Detail & Related papers  (2024-03-10T04:38:27Z)
• Weakly Supervised 3D Object Detection via Multi-Level Visual Guidance [72.6809373191638]
  We propose a framework to study how to leverage constraints between 2D and 3D domains without requiring any 3D labels. Specifically, we design a feature-level constraint to align LiDAR and image features based on object-aware regions. Second, the output-level constraint is developed to enforce the overlap between 2D and projected 3D box estimations. Third, the training-level constraint is utilized by producing accurate and consistent 3D pseudo-labels that align with the visual data.
  arXiv  Detail & Related papers  (2023-12-12T18:57:25Z)
• Tracking Objects with 3D Representation from Videos [57.641129788552675]
  We propose a new 2D Multiple Object Tracking paradigm, called P3DTrack. With 3D object representation learning from Pseudo 3D object labels in monocular videos, we propose a new 2D MOT paradigm, called P3DTrack.
  arXiv  Detail & Related papers  (2023-06-08T17:58:45Z)
• Object as Query: Lifting any 2D Object Detector to 3D Detection [30.393111518104313]
  We design Multi-View 2D Objects guided 3D Object Detector (MV2D) MV2D exploits 2D detectors to generate object queries conditioned on the rich image semantics. For the generated queries, we design a sparse cross attention module to force them to focus on the features of specific objects.
  arXiv  Detail & Related papers  (2023-01-06T04:08:20Z)
• Homography Loss for Monocular 3D Object Detection [54.04870007473932]
  A differentiable loss function, termed as Homography Loss, is proposed to achieve the goal, which exploits both 2D and 3D information. Our method yields the best performance compared with the other state-of-the-arts by a large margin on KITTI 3D datasets.
  arXiv  Detail & Related papers  (2022-04-02T03:48:03Z)
• YOLOStereo3D: A Step Back to 2D for Efficient Stereo 3D Detection [6.5702792909006735]
  YOLOStereo3D is trained on one single GPU and runs at more than ten fps. It demonstrates performance comparable to state-of-the-art stereo 3D detection frameworks without usage of LiDAR data.
  arXiv  Detail & Related papers  (2021-03-17T03:43:54Z)
• DSGN: Deep Stereo Geometry Network for 3D Object Detection [79.16397166985706]
  There is a large performance gap between image-based and LiDAR-based 3D object detectors. Our method, called Deep Stereo Geometry Network (DSGN), significantly reduces this gap. For the first time, we provide a simple and effective one-stage stereo-based 3D detection pipeline.
  arXiv  Detail & Related papers  (2020-01-10T11:44:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.