Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Leijenaar, Remco F., Kasaei, Hamidreza
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908425270591488
author Leijenaar, Remco F.
Kasaei, Hamidreza
author_facet Leijenaar, Remco F.
Kasaei, Hamidreza
contents Learning semantically meaningful representations from unstructured 3D point clouds remains a central challenge in computer vision, especially in the absence of large-scale labeled datasets. While masked point modeling (MPM) is widely used in self-supervised 3D learning, its reconstruction-based objective can limit its ability to capture high-level semantics. We propose AsymDSD, an Asymmetric Dual Self-Distillation framework that unifies masked modeling and invariance learning through prediction in the latent space rather than the input space. AsymDSD builds on a joint embedding architecture and introduces several key design choices: an efficient asymmetric setup, disabling attention between masked queries to prevent shape leakage, multi-mask sampling, and a point cloud adaptation of multi-crop. AsymDSD achieves state-of-the-art results on ScanObjectNN (90.53%) and further improves to 93.72% when pretrained on 930k shapes, surpassing prior methods.
format Preprint
id arxiv_https___arxiv_org_abs_2506_21724
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning
Leijenaar, Remco F.
Kasaei, Hamidreza
Computer Vision and Pattern Recognition
Learning semantically meaningful representations from unstructured 3D point clouds remains a central challenge in computer vision, especially in the absence of large-scale labeled datasets. While masked point modeling (MPM) is widely used in self-supervised 3D learning, its reconstruction-based objective can limit its ability to capture high-level semantics. We propose AsymDSD, an Asymmetric Dual Self-Distillation framework that unifies masked modeling and invariance learning through prediction in the latent space rather than the input space. AsymDSD builds on a joint embedding architecture and introduces several key design choices: an efficient asymmetric setup, disabling attention between masked queries to prevent shape leakage, multi-mask sampling, and a point cloud adaptation of multi-crop. AsymDSD achieves state-of-the-art results on ScanObjectNN (90.53%) and further improves to 93.72% when pretrained on 930k shapes, surpassing prior methods.
title Asymmetric Dual Self-Distillation for 3D Self-Supervised Representation Learning
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.21724