EUVIMEDEUVIMED entdecken
Crossref

Diffusion-Driven Self-Supervised Learning for Shape Reconstruction and Pose Estimation

Jingtao Sun, Yaonan Wang, Mingtao Feng, Chao Ding, Mike Zheng Shou, Ajmal Mian

IEEE Transactions on Pattern Analysis and Machine Intelligence · 2026-05 · 48 (5) · 5113-5130

DOI: 10.1109/tpami.2025.3647855 ↗

Kurz gefasst

Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2026-05. Verfasst von Jingtao Sun und 5 weiteren Autor:innen. Die Arbeit wurde bisher 2-mal zitiert (Crossref) und nennt 78 Literaturangaben. Gefördert von: National Research Foundation Singapore, Australian Research Council Future Fellowship.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Fully-supervised category-level pose estimation aims to determine the 6-DoF poses of unseen instances from known categories, requiring expensive manual labeling costs. Recently, various self-supervised category-level pose estimation methods have been proposed to reduce the requirement of the annotated datasets. However, most methods rely on synthetic data or 3D CAD model, and they are typically limited to addressing single-object pose problems without considering multi-objective tasks or shape reconstruction. To overcome these challenges and limitations, we introduce a diffusion-driven self-supervised network for multi-object shape reconstruction and categorical pose estimation, only leveraging the shape priors. Specifically, to capture the SE(3)-equivariant pose features and 3D scale-invariant shape information, we present a Prior-Aware Pyramid 3D Point Transformer. This module adopts a point convolutional layer with radial-kernels for pose-aware learning and a 3D scale-invariant graph convolution layer for object-level shape representation. Furthermore, we introduce a Pretrain-to-Refine Self-Supervised Training Paradigm to train our network. It enables proposed network to capture the associations between shape priors and observations, addressing the challenge of intra-class shape variations by utilising the diffusion mechanism. Extensive experiments conducted on four public datasets and a self-built dataset demonstrate that our method significantly outperforms state-of-the-art self-supervised category-level baselines and even surpasses some fully-supervised instance-level and category-level methods. The project page is released at Self-SRPE.

Förderung

  • National Research Foundation Singapore
  • Australian Research Council Future Fellowship

Zitieren

APA

Sun, J., Wang, Y., Feng, M., Ding, C., Shou, M. Z., & Mian, A. (2026). Diffusion-Driven Self-Supervised Learning for Shape Reconstruction and Pose Estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 48(5), 5113-5130. https://doi.org/10.1109/tpami.2025.3647855

Vancouver

Sun J, Wang Y, Feng M, Ding C, Shou MZ, Mian A. Diffusion-Driven Self-Supervised Learning for Shape Reconstruction and Pose Estimation. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2026;48(5):5113-5130. doi:10.1109/tpami.2025.3647855

BibTeX
@article{Sun2026,
  author = {Sun, Jingtao and Wang, Yaonan and Feng, Mingtao and Ding, Chao and Shou, Mike Zheng and Mian, Ajmal},
  title = {Diffusion-Driven Self-Supervised Learning for Shape Reconstruction and Pose Estimation},
  journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year = {2026},
  volume = {48},
  number = {5},
  pages = {5113--5130},
  doi = {10.1109/tpami.2025.3647855}
}
Kostenlos starten →