Enhancing LiDAR Point Features with Foundation Model Priors for 3D Object Detection

2025-07-18 · Yujian Mo, Yan Wu, Junqiao Zhao, Jijun Wang, Yinghao Hu, Jun Yan arxiv

Recent advances in foundation models have opened up new possibilities for enhancing 3D perception. In particular, DepthAnything offers dense and reliable geometric priors from monocular RGB images, which can complement sparse LiDAR data in autonomous driving scenarios. However, such priors remain underutilized in LiDAR-based 3D object detection. In this paper, we address the limited expressiveness of raw LiDAR point features, especially the weak discriminative capability of the reflectance attribute, by introducing depth priors predicted by DepthAnything. These priors are fused with the original LiDAR attributes to enrich each point's representation. To leverage the enhanced point features, we propose a point-wise feature extraction module. Then, a Dual-Path RoI feature extraction framework is employed, comprising a voxel-based branch for global semantic context and a point-based branch for fine-grained structural details. To effectively integrate the complementary RoI features, we introduce a bidirectional gated RoI feature fusion module that balances global and local cues. Extensive experiments on the KITTI benchmark show that our method consistently improves detection accuracy, demonstrating the value of incorporating visual foundation model priors into LiDAR-based 3D object detection.

📄 PDF Abstract BibTeX arXiv:2507.13899

Code (0)

등록된 구현이 없습니다.

Tasks

3D Object Detection Autonomous Driving

Geometry Without Coordinates: LiDAR Diffusion as a 3D Feature Bridge

2026-09-09 · Samed Doğan, Nico Leuze, Alfred Schöttl arxiv

Transferring the rich priors of large 2D foundation models to sparse 3D LiDAR remains challenging, as training native 3D foundation models at comparable scale is limited by data and annotation scarcity. We introduce a Li…

Semantic Segmentation

Sparse-LiDAR Prompting of Monocular Geometry Foundations: An Empirical Study Toward Long-Range Driving Depth

2026-05-26 · Kai Zheng, Qiang Feng, Xingjian Liu, Wenquan Tan 외 arxiv

Sparse-LiDAR-prompted depth foundation models (PromptDA, Prior Depth Anything, DMD3C) have shown strong results on indoor scenes or within KITTI's standard 80-meter evaluation cap. However, two limitations remain: (i) sy…

R3DPA: Leveraging 3D Representation Alignment and RGB Pretrained Priors for LiDAR Scene Generation

2026-01-12 · Nicolas Sereyjol-Garros, Ellington Kirby, Victor Besnier, Nermin Samet arxiv

LiDAR scene synthesis is an emerging solution to scarcity in 3D data for robotic tasks such as autonomous driving. Recent approaches employ diffusion or flow matching models to generate realistic scenes, but 3D data rema…

Autonomous Driving Scene Generation Point Clouds

D$^2$GS: Dense Depth Regularization for LiDAR-free Urban Scene Reconstruction

2025-10-29 · Kejing Xia, Jidong Jia, Ke Jin, Yucai Bai 외 arxiv

Recently, Gaussian Splatting (GS) has shown great potential for urban scene reconstruction in the field of autonomous driving. However, current urban scene reconstruction methods often depend on multimodal sensors as inp…

Autonomous Driving Point Clouds

TOLiD: Bridging the Architecture Gap in Vision Foundation Model to LiDAR Pretraining via Token Lifting for Distillation

2026-07-12 · Sutharsan Mahendran, Darshana Priyasad, Kaushik Roy, Tharindu Fernando 외 arxiv

Cross-modal distillation from Vision Foundation Models (VFMs) to LiDAR backbones has recently emerged as a self-supervised pretraining strategy that reduces reliance on dense point-wise annotation for 3D scene understand…

Representation Learning Scene Understanding