4D Contrastive Superflows are Dense 3D Representation Learners
- URL: http://arxiv.org/abs/2407.06190v2
- Date: Wed, 10 Jul 2024 01:32:28 GMT
- Title: 4D Contrastive Superflows are Dense 3D Representation Learners
- Authors: Xiang Xu, Lingdong Kong, Hui Shuai, Wenwei Zhang, Liang Pan, Kai Chen, Ziwei Liu, Qingshan Liu,
- Abstract summary: 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.
- Score: 62.433137130087445
- License: http://creativecommons.org/licenses/by-sa/4.0/
- Abstract: In the realm of autonomous driving, accurate 3D perception is the foundation. However, developing such models relies on extensive human annotations -- a process that is both costly and labor-intensive. To address this challenge from a data representation learning perspective, we introduce SuperFlow, a novel framework designed to harness consecutive LiDAR-camera pairs for establishing spatiotemporal pretraining objectives. SuperFlow stands out by integrating two key designs: 1) a dense-to-sparse consistency regularization, which promotes insensitivity to point cloud density variations during feature learning, and 2) a flow-based contrastive learning module, carefully crafted to extract meaningful temporal cues from readily available sensor calibrations. To further boost learning efficiency, we incorporate a plug-and-play view consistency module that enhances the alignment of the knowledge distilled from camera views. Extensive comparative and ablation studies across 11 heterogeneous LiDAR datasets validate our effectiveness and superiority. Additionally, we observe several interesting emerging properties by scaling up the 2D and 3D backbones during pretraining, shedding light on the future research of 3D foundation models for LiDAR-based perception.
Related papers
- LargeAD: Large-Scale Cross-Sensor Data Pretraining for Autonomous Driving [52.83707400688378]
LargeAD is a versatile and scalable framework designed for large-scale 3D pretraining across diverse real-world driving datasets.
Our framework leverages VFMs to extract semantically rich superpixels from 2D images, which are aligned with LiDAR point clouds to generate high-quality contrastive samples.
Our approach delivers significant performance improvements over state-of-the-art methods in both linear probing and fine-tuning tasks for both LiDAR-based segmentation and object detection.
arXiv Detail & Related papers (2025-01-07T18:59:59Z) - A Lesson in Splats: Teacher-Guided Diffusion for 3D Gaussian Splats Generation with 2D Supervision [65.33043028101471]
We introduce a diffusion model for Gaussian Splats, SplatDiffusion, to enable generation of three-dimensional structures from single images.
Existing methods rely on deterministic, feed-forward predictions, which limit their ability to handle the inherent ambiguity of 3D inference from 2D data.
arXiv Detail & Related papers (2024-12-01T00:29:57Z) - Unleashing the Potential of Mamba: Boosting a LiDAR 3D Sparse Detector by Using Cross-Model Knowledge Distillation [22.653014803666668]
We propose a Faster LiDAR 3D object detection framework, called FASD, which implements heterogeneous model distillation by adaptively uniform cross-model voxel features.
We aim to distill the transformer's capacity for high-performance sequence modeling into Mamba models with low FLOPs, achieving a significant improvement in accuracy through knowledge transfer.
We evaluated the framework on datasets and nuScenes, achieving a 4x reduction in resource consumption and a 1-2% performance improvement over the current SoTA methods.
arXiv Detail & Related papers (2024-09-17T09:30:43Z) - 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.
Results demonstrate that LaserMix++ outperforms fully supervised alternatives, achieving comparable accuracy with five times fewer annotations.
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) - SPOT: Scalable 3D Pre-training via Occupancy Prediction for Learning Transferable 3D Representations [76.45009891152178]
Pretraining-finetuning approach can alleviate the labeling burden by fine-tuning a pre-trained backbone across various downstream datasets as well as tasks.
We show, for the first time, that general representations learning can be achieved through the task of occupancy prediction.
Our findings will facilitate the understanding of LiDAR points and pave the way for future advancements in LiDAR pre-training.
arXiv Detail & Related papers (2023-09-19T11:13:01Z) - Spatio-temporal Self-Supervised Representation Learning for 3D Point
Clouds [96.9027094562957]
We introduce a-temporal representation learning framework, capable of learning from unlabeled tasks.
Inspired by how infants learn from visual data in the wild, we explore rich cues derived from the 3D data.
STRL takes two temporally-related frames from a 3D point cloud sequence as the input, transforms it with the spatial data augmentation, and learns the invariant representation self-supervisedly.
arXiv Detail & Related papers (2021-09-01T04:17:11Z)
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.