Self-supervised Learning of Rotation-invariant 3D Point Set Features using Transformer and its Self-distillation

• URL: http://arxiv.org/abs/2308.04725v2
• Date: Thu, 18 Apr 2024 23:40:16 GMT
• Title: Self-supervised Learning of Rotation-invariant 3D Point Set Features using Transformer and its Self-distillation
• Authors: Takahiko Furuya, Zhoujie Chen, Ryutarou Ohbuchi, Zhenzhong Kuang,
• Abstract summary: This paper proposes a novel self-supervised learning framework for acquiring accurate and rotation-invariant 3D point set features at object-level. We employ a self-attention mechanism to refine the tokens and aggregate them into an expressive rotation-invariant feature per 3D point set. Our proposed algorithm learns rotation-invariant 3D point set features that are more accurate than those learned by existing algorithms.
• Score: 3.1652399282742536
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Invariance against rotations of 3D objects is an important property in analyzing 3D point set data. Conventional 3D point set DNNs having rotation invariance typically obtain accurate 3D shape features via supervised learning by using labeled 3D point sets as training samples. However, due to the rapid increase in 3D point set data and the high cost of labeling, a framework to learn rotation-invariant 3D shape features from numerous unlabeled 3D point sets is required. This paper proposes a novel self-supervised learning framework for acquiring accurate and rotation-invariant 3D point set features at object-level. Our proposed lightweight DNN architecture decomposes an input 3D point set into multiple global-scale regions, called tokens, that preserve the spatial layout of partial shapes composing the 3D object. We employ a self-attention mechanism to refine the tokens and aggregate them into an expressive rotation-invariant feature per 3D point set. Our DNN is effectively trained by using pseudo-labels generated by a self-distillation framework. To facilitate the learning of accurate features, we propose to combine multi-crop and cut-mix data augmentation techniques to diversify 3D point sets for training. Through a comprehensive evaluation, we empirically demonstrate that, (1) existing rotation-invariant DNN architectures designed for supervised learning do not necessarily learn accurate 3D shape features under a self-supervised learning scenario, and (2) our proposed algorithm learns rotation-invariant 3D point set features that are more accurate than those learned by existing algorithms. Code is available at https://github.com/takahikof/RIPT_SDMM

Related papers

• Point-DETR3D: Leveraging Imagery Data with Spatial Point Prior for Weakly Semi-supervised 3D Object Detection [32.86369670395974]
  We introduce Point-DETR3D, a teacher-student framework for weakly semi-supervised 3D detection. With only 5% of labeled data, Point-DETR3D achieves over 90% performance of its fully supervised counterpart.
  arXiv  Detail & Related papers  (2024-03-22T16:11:29Z)
• MaskLRF: Self-supervised Pretraining via Masked Autoencoding of Local Reference Frames for Rotation-invariant 3D Point Set Analysis [1.19658449368018]
  This paper develops, for the first time, a rotation-invariant self-supervised pretraining framework for practical 3D point set analysis. The proposed algorithm, called MaskLRF, learns rotation-invariant and highly generalizable latent features via masked autoencoding of 3D points. I confirm that MaskLRF achieves new state-of-the-art accuracies in analyzing 3D point sets having inconsistent orientations.
  arXiv  Detail & Related papers  (2024-03-01T00:42:49Z)
• Weakly Supervised 3D Object Detection via Multi-Level Visual Guidance [72.6809373191638]
  We propose a framework to study how to leverage constraints between 2D and 3D domains without requiring any 3D labels. Specifically, we design a feature-level constraint to align LiDAR and image features based on object-aware regions. Second, the output-level constraint is developed to enforce the overlap between 2D and projected 3D box estimations. Third, the training-level constraint is utilized by producing accurate and consistent 3D pseudo-labels that align with the visual data.
  arXiv  Detail & Related papers  (2023-12-12T18:57:25Z)
• SNAKE: Shape-aware Neural 3D Keypoint Field [62.91169625183118]
  Detecting 3D keypoints from point clouds is important for shape reconstruction. This work investigates the dual question: can shape reconstruction benefit 3D keypoint detection? We propose a novel unsupervised paradigm named SNAKE, which is short for shape-aware neural 3D keypoint field.
  arXiv  Detail & Related papers  (2022-06-03T17:58:43Z)
• ConDor: Self-Supervised Canonicalization of 3D Pose for Partial Shapes [55.689763519293464]
  ConDor is a self-supervised method that learns to canonicalize the 3D orientation and position for full and partial 3D point clouds. During inference, our method takes an unseen full or partial 3D point cloud at an arbitrary pose and outputs an equivariant canonical pose.
  arXiv  Detail & Related papers  (2022-01-19T18:57:21Z)
• ST3D: Self-training for Unsupervised Domain Adaptation on 3D ObjectDetection [78.71826145162092]
  We present a new domain adaptive self-training pipeline, named ST3D, for unsupervised domain adaptation on 3D object detection from point clouds. Our ST3D achieves state-of-the-art performance on all evaluated datasets and even surpasses fully supervised results on KITTI 3D object detection benchmark.
  arXiv  Detail & Related papers  (2021-03-09T10:51:24Z)
• Adjoint Rigid Transform Network: Task-conditioned Alignment of 3D Shapes [86.2129580231191]
  Adjoint Rigid Transform (ART) Network is a neural module which can be integrated with a variety of 3D networks. ART learns to rotate input shapes to a learned canonical orientation, which is crucial for a lot of tasks. We will release our code and pre-trained models for further research.
  arXiv  Detail & Related papers  (2021-02-01T20:58:45Z)
• Rotation-Invariant Local-to-Global Representation Learning for 3D Point Cloud [42.86112554931754]
  We propose a local-to-global representation learning algorithm for 3D point cloud data. Our model takes advantage of multi-level abstraction based on graph convolutional neural networks. The proposed algorithm presents the state-of-the-art performance on the rotation-augmented 3D object recognition and segmentation benchmarks.
  arXiv  Detail & Related papers  (2020-10-07T10:30:20Z)
• D3Feat: Joint Learning of Dense Detection and Description of 3D Local Features [51.04841465193678]
  We leverage a 3D fully convolutional network for 3D point clouds. We propose a novel and practical learning mechanism that densely predicts both a detection score and a description feature for each 3D point. Our method achieves state-of-the-art results in both indoor and outdoor scenarios.
  arXiv  Detail & Related papers  (2020-03-06T12:51:09Z)

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.