E3D: Enhancing Sparsely-Supervised 3D Object Detector with Large Multimodal Models
Keywords: sparsely-supervised object detection, point cloud, large multimodal model
Abstract: Recently, sparsely-supervised 3D object detection has gained great attention, achieving performance close to that of fully-supervised 3D objectors with only a few annotated instances. Nevertheless, these methods suffer challenges when the accurate labels are extremely limited. In this paper, we propose an Ehanced 3D object Detection strategy, termed E3D, explicitly utilizing the prior knowledge from Large Multimodal Models (LMMs) to enhance the feature discrimination capability of the 3D detector under sparse annotation settings. Specifically, we first develop a Confident Points Semantic Transfer (CPST) module that generates high-quality seed points through boundary-constrained center cluster selection. Based on these seed points, we introduce a Dynamic Cluster Pseudo-label Generation (DCPG) module that yields pseudo-supervision signals from the geometry shape of multi-scale neighbor points. Additionally, we design a Distribution Shape score (DS score) that chooses high-quality supervision signals for the initial training of the 3D detector. By utilizing E3D, existing leading sparsely-supervised CoIn++ is improved by an average of 11.63% under the annotation rate of 2%. Moreover, we have verified our E3D in the zero-shot setting, and the results demonstrate its performance exceeding that of the state-of-the-art methods. The code will be made publicly available.
Primary Area: applications to computer vision, audio, language, and other modalities
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2025/AuthorGuide.
Reciprocal Reviewing: I understand the reciprocal reviewing requirement as described on https://iclr.cc/Conferences/2025/CallForPapers. If none of the authors are registered as a reviewer, it may result in a desk rejection at the discretion of the program chairs. To request an exception, please complete this form at https://forms.gle/Huojr6VjkFxiQsUp6.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors’ identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 10702
Loading