About

LiMA is a novel long-term image-to-LiDAR Memory Aggregation framework that explicitly captures longer range temporal correlations to enhance LiDAR representation learning. It 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.

:memo: Updates

• [2025.06] - Our paper LiMA has been accepted to ICCV 2025! :tada:

Table of Content

• Installation
• Data Preparation
• Getting Started
• Main Results
• License
• Citation
• Acknowledgments

:gear: Installation

For details related to installation and environment setups, kindly refer to INSTALL.md.

:hotsprings: Data Preparation

Kindly refer to DATA_PREPAER.md for the details to prepare the datasets.

:rocket: Getting Started

To learn more usage about this codebase, kindly refer to GET_STARTED.md.

:bar_chart: Main Results

Comparisons of State-of-the-Art Pretraining Methods

Domain Generalization Study

3D Object Detection

Cosine Similarity

Cosine similarity between a query point (marked as the red dot) and: (1) image features, and (2) LiDAR point features projected onto the image. Colors range from red (indicating high similarity) to blue (indicating low similarity). Best viewed in colors.

Qualitative Assessment

Qualitative assessments of state-of-the-art methods, pretrained on nuScenes and fine-tuned on nuScenes with 1% annotations. The error maps depict correct and incorrect predictions in gray and red, respectively. Best viewed in colors.

Qualitative assessments of object detection, pretrained on nuScenes and fine-tuned on nuScenes with 5% annotations. The groundtruth/predicted results are highlighted with blue/red boxes, respectively. Best viewed in colors.

License

This work is under the Apache License Version 2.0, while some specific implementations in this codebase might be with other licenses.

Kindly refer to LICENSE.md for a more careful check, if you are using our code for commercial matters.

Citation

If you find this work helpful for your research, please kindly consider citing our paper:

@inproceedings{xu2025lima,
    title = {Beyond One Shot, Beyond One Perspective: Cross-View and Long-Horizon Distillation for Better LiDAR Representations},
    author = {Xu, Xiang and Kong, Lingdong and Wang, Song and Zhou, Chuanwei and Liu, Qingshan},
    booktitle = {IEEE/CVF International Conference on Computer Vision},
    year = {2025}
}

Acknowledgments

This work is developed based on the MMDetection3D codebase.

MMDetection3D is an open-source object detection toolbox based on PyTorch, towards the next-generation platform for general 3D perception. It is a part of the OpenMMLab project developed by MMLab.

We acknowledge the use of the following public resources during the couuse of this work: ¹nuScenes, ²nuScenes-devkit, ³SemanticKITTI, ⁴SemanticKITTI-API, , ⁵WaymoOpenDataset, ⁶Synth4D, ⁷ScribbleKITTI, ⁸RELLIS-3D, ⁹SemanticPOSS, ¹⁰SemanticSTF, ¹¹SynthLiDAR, ¹²DAPS-3D, ¹³Robo3D, ¹⁴SLidR, ¹⁵DINOv2, ¹⁶FRNet, ¹⁷SuperFlow, ¹⁸torchsparse, ¹⁹ScaLR. :heart_decoration: