3D Unsupervised Learning by Distilling 2D Open-Vocabulary Segmentation Models for Autonomous Driving

• URL: http://arxiv.org/abs/2405.15286v2
• Date: Sat, 21 Sep 2024 07:25:40 GMT
• Title: 3D Unsupervised Learning by Distilling 2D Open-Vocabulary Segmentation Models for Autonomous Driving
• Authors: Boyi Sun, Yuhang Liu, Xingxia Wang, Bin Tian, Long Chen, Fei-Yue Wang,
• Abstract summary: We propose UOV, a novel 3D Unsupervised framework assisted by 2D Open-Vocabulary segmentation models. In the first stage, we innovatively integrate high-quality textual and image features of 2D open-vocabulary models. In the second stage, spatial mapping between point clouds and images is utilized to generate pseudo-labels.
• Score: 17.42913935045091
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Point cloud data labeling is considered a time-consuming and expensive task in autonomous driving, whereas unsupervised learning can avoid it by learning point cloud representations from unannotated data. In this paper, we propose UOV, a novel 3D Unsupervised framework assisted by 2D Open-Vocabulary segmentation models. It consists of two stages: In the first stage, we innovatively integrate high-quality textual and image features of 2D open-vocabulary models and propose the Tri-Modal contrastive Pre-training (TMP). In the second stage, spatial mapping between point clouds and images is utilized to generate pseudo-labels, enabling cross-modal knowledge distillation. Besides, we introduce the Approximate Flat Interaction (AFI) to address the noise during alignment and label confusion. To validate the superiority of UOV, extensive experiments are conducted on multiple related datasets. We achieved a record-breaking 47.73% mIoU on the annotation-free point cloud segmentation task in nuScenes, surpassing the previous best model by 10.70% mIoU. Meanwhile, the performance of fine-tuning with 1% data on nuScenes and SemanticKITTI reached a remarkable 51.75% mIoU and 48.14% mIoU, outperforming all previous pre-trained models.

Related papers

• 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)
• Leveraging Large-Scale Pretrained Vision Foundation Models for Label-Efficient 3D Point Cloud Segmentation [67.07112533415116]
  We present a novel framework that adapts various foundational models for the 3D point cloud segmentation task. Our approach involves making initial predictions of 2D semantic masks using different large vision models. To generate robust 3D semantic pseudo labels, we introduce a semantic label fusion strategy that effectively combines all the results via voting.
  arXiv  Detail & Related papers  (2023-11-03T15:41:15Z)
• Segment Any Point Cloud Sequences by Distilling Vision Foundation Models [55.12618600523729]
  Seal is a framework that harnesses vision foundation models (VFMs) for segmenting diverse automotive point cloud sequences. Seal exhibits three appealing properties: Scalability, consistency and generalizability.
  arXiv  Detail & Related papers  (2023-06-15T17:59:54Z)
• Point2Vec for Self-Supervised Representation Learning on Point Clouds [66.53955515020053]
  We extend data2vec to the point cloud domain and report encouraging results on several downstream tasks. We propose point2vec, which unleashes the full potential of data2vec-like pre-training on point clouds.
  arXiv  Detail & Related papers  (2023-03-29T10:08:29Z)
• PointVST: Self-Supervised Pre-training for 3D Point Clouds via View-Specific Point-to-Image Translation [64.858505571083]
  This paper proposes a translative pre-training framework, namely PointVST. It is driven by a novel self-supervised pretext task of cross-modal translation from 3D point clouds to their corresponding diverse forms of 2D rendered images.
  arXiv  Detail & Related papers  (2022-12-29T07:03:29Z)
• 3D Point Cloud Pre-training with Knowledge Distillation from 2D Images [128.40422211090078]
  We propose a knowledge distillation method for 3D point cloud pre-trained models to acquire knowledge directly from the 2D representation learning model. Specifically, we introduce a cross-attention mechanism to extract concept features from 3D point cloud and compare them with the semantic information from 2D images. In this scheme, the point cloud pre-trained models learn directly from rich information contained in 2D teacher models.
  arXiv  Detail & Related papers  (2022-12-17T23:21:04Z)
• Efficient Urban-scale Point Clouds Segmentation with BEV Projection [0.0]
  Most deep point clouds models directly conduct learning on 3D point clouds. We propose to transfer the 3D point clouds to dense bird's-eye-view projection.
  arXiv  Detail & Related papers  (2021-09-19T06:49:59Z)

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.