Beyond One Shot, Beyond One Perspective: Cross-View and Long-Horizon Distillation for Better LiDAR Representations

• URL: http://arxiv.org/abs/2507.05260v1
• Date: Mon, 07 Jul 2025 17:59:58 GMT
• Title: Beyond One Shot, Beyond One Perspective: Cross-View and Long-Horizon Distillation for Better LiDAR Representations
• Authors: Xiang Xu, Lingdong Kong, Song Wang, Chuanwei Zhou, Qingshan Liu,
• Abstract summary: LiMA is a novel framework that captures longer range temporal correlations to enhance LiDAR representation learning.<n>LiMA has high pretraining efficiency and incurs no additional computational overhead during downstream tasks.<n> experiments on mainstream LiDAR-based perception benchmarks demonstrate that LiMA significantly improves both LiDAR semantic segmentation and 3D object detection.
• Score: 23.21118045286231
• License: http://creativecommons.org/licenses/by-sa/4.0/
• Abstract: LiDAR representation learning aims to extract rich structural and semantic information from large-scale, readily available datasets, reducing reliance on costly human annotations. However, existing LiDAR representation strategies often overlook the inherent spatiotemporal cues in LiDAR sequences, limiting their effectiveness. In this work, we propose LiMA, a novel long-term image-to-LiDAR Memory Aggregation framework that explicitly captures longer range temporal correlations to enhance LiDAR representation learning. LiMA comprises three key components: 1) a Cross-View Aggregation module that aligns and fuses overlapping regions across neighboring camera views, constructing a more unified and redundancy-free memory bank; 2) a Long-Term Feature Propagation mechanism that efficiently aligns and integrates multi-frame image features, reinforcing temporal coherence during LiDAR representation learning; and 3) a Cross-Sequence Memory Alignment strategy that enforces consistency across driving sequences, improving generalization to unseen environments. LiMA maintains high pretraining efficiency and incurs no additional computational overhead during downstream tasks. Extensive experiments on mainstream LiDAR-based perception benchmarks demonstrate that LiMA significantly improves both LiDAR semantic segmentation and 3D object detection. We hope this work inspires more effective pretraining paradigms for autonomous driving. The code has be made publicly accessible for future research.

Related papers

• SuperFlow++: Enhanced Spatiotemporal Consistency for Cross-Modal Data Pretraining [62.433137130087445]
  SuperFlow++ is a novel framework that integrates pretraining and downstream tasks using consecutive camera pairs.<n>We show that SuperFlow++ outperforms state-of-the-art methods across diverse tasks and driving conditions.<n>With strong generalizability and computational efficiency, SuperFlow++ establishes a new benchmark for data-efficient LiDAR-based perception in autonomous driving.
  arXiv  Detail & Related papers  (2025-03-25T17:59:57Z)
• LiMoE: Mixture of LiDAR Representation Learners from Automotive Scenes [55.33167217384738]
  LiMoE is a framework that integrates the Mixture of Experts (MoE) paradigm into LiDAR data representation learning.<n>Our approach consists of three stages: Image-to-LiDAR Pretraining, Contrastive Mixture Learning (CML), and Semantic Mixture Supervision (SMS)
  arXiv  Detail & Related papers  (2025-01-07T18:59:58Z)
• LiDAR-GS:Real-time LiDAR Re-Simulation using Gaussian Splatting [50.808933338389686]
  We present LiDAR-GS, a real-time, high-fidelity re-simulation of LiDAR scans in public urban road scenes.<n>The method achieves state-of-the-art results in both rendering frame rate and quality on publically available large scene datasets.
  arXiv  Detail & Related papers  (2024-10-07T15:07:56Z)
• TASeg: Temporal Aggregation Network for LiDAR Semantic Segmentation [80.13343299606146]
  We propose a Temporal LiDAR Aggregation and Distillation (TLAD) algorithm, which leverages historical priors to assign different aggregation steps for different classes. To make full use of temporal images, we design a Temporal Image Aggregation and Fusion (TIAF) module, which can greatly expand the camera FOV. We also develop a Static-Moving Switch Augmentation (SMSA) algorithm, which utilizes sufficient temporal information to enable objects to switch their motion states freely.
  arXiv  Detail & Related papers  (2024-07-13T03:00:16Z)
• 4D Contrastive Superflows are Dense 3D Representation Learners [62.433137130087445]
  We introduce SuperFlow, a novel framework designed to harness consecutive LiDAR-camera pairs for establishing pretraining objectives. To further boost learning efficiency, we incorporate a plug-and-play view consistency module that enhances alignment of the knowledge distilled from camera views.
  arXiv  Detail & Related papers  (2024-07-08T17:59:54Z)
• Multi-Modal Data-Efficient 3D Scene Understanding for Autonomous Driving [58.16024314532443]
  We introduce LaserMix++, a framework that integrates laser beam manipulations from disparate LiDAR scans and incorporates LiDAR-camera correspondences to assist data-efficient learning.<n>Results demonstrate that LaserMix++ outperforms fully supervised alternatives, achieving comparable accuracy with five times fewer annotations.<n>This substantial advancement underscores the potential of semi-supervised approaches in reducing the reliance on extensive labeled data in LiDAR-based 3D scene understanding systems.
  arXiv  Detail & Related papers  (2024-05-08T17:59:53Z)
• Efficient Spatial-Temporal Information Fusion for LiDAR-Based 3D Moving Object Segmentation [23.666607237164186]
  We propose a novel deep neural network exploiting both spatial-temporal information and different representation modalities of LiDAR scans to improve LiDAR-MOS performance. Specifically, we first use a range image-based dual-branch structure to separately deal with spatial and temporal information. We also use a point refinement module via 3D sparse convolution to fuse the information from both LiDAR range image and point cloud representations.
  arXiv  Detail & Related papers  (2022-07-05T17:59:17Z)
• EfficientLPS: Efficient LiDAR Panoptic Segmentation [30.249379810530165]
  We present the novel Efficient LiDAR Panoptic architecture that addresses multiple challenges in segmenting LiDAR point clouds. EfficientLPS comprises of a novel shared backbone that encodes with strengthened geometric transformation modeling capacity. We benchmark our proposed model on two large-scale LiDAR datasets.
  arXiv  Detail & Related papers  (2021-02-16T08:14:52Z)

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.