SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds

Ali Bahri, Moslem Yazdanpanah, Sahar Dastani, Mehrdad Noori, Gustavo Adolfo Vargas Hakim, David Osowiechi, Farzad Beizaee, Ismail Ben Ayed, Christian Desrosiers
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:2442-2457, 2025.

Abstract

Test-Time Training has emerged as a promising solution to address distribution shifts in 3D point cloud classification. However, existing methods often rely on computationally expensive backpropagation during adaptation, limiting their applicability in real-world, time-sensitive scenarios. In this paper, we introduce SMART-PC, a skeleton-based framework that enhances resilience to corruptions by leveraging the geometric structure of 3D point clouds. During pre-training, our method predicts skeletal representations, enabling the model to extract robust and meaningful geometric features that are less sensitive to corruptions, thereby improving adaptability to test-time distribution shifts. Unlike prior approaches, SMART-PC achieves real-time adaptation by eliminating backpropagation and updating only BatchNorm statistics, resulting in a lightweight and efficient framework capable of achieving high frame-per-second rates while maintaining superior classification performance. Extensive experiments on benchmark datasets, including ModelNet40-C, ShapeNet-C, and ScanObjectNN-C, demonstrate that SMART-PC achieves state-of-the-art results, outperforming existing methods such as MATE in terms of both accuracy and computational efficiency. The implementation is available at: https://github.com/AliBahri94/SMART-PC.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-bahri25a, title = {{SMART}-{PC}: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds}, author = {Bahri, Ali and Yazdanpanah, Moslem and Dastani, Sahar and Noori, Mehrdad and Vargas Hakim, Gustavo Adolfo and Osowiechi, David and Beizaee, Farzad and Ayed, Ismail Ben and Desrosiers, Christian}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {2442--2457}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/bahri25a/bahri25a.pdf}, url = {https://proceedings.mlr.press/v267/bahri25a.html}, abstract = {Test-Time Training has emerged as a promising solution to address distribution shifts in 3D point cloud classification. However, existing methods often rely on computationally expensive backpropagation during adaptation, limiting their applicability in real-world, time-sensitive scenarios. In this paper, we introduce SMART-PC, a skeleton-based framework that enhances resilience to corruptions by leveraging the geometric structure of 3D point clouds. During pre-training, our method predicts skeletal representations, enabling the model to extract robust and meaningful geometric features that are less sensitive to corruptions, thereby improving adaptability to test-time distribution shifts. Unlike prior approaches, SMART-PC achieves real-time adaptation by eliminating backpropagation and updating only BatchNorm statistics, resulting in a lightweight and efficient framework capable of achieving high frame-per-second rates while maintaining superior classification performance. Extensive experiments on benchmark datasets, including ModelNet40-C, ShapeNet-C, and ScanObjectNN-C, demonstrate that SMART-PC achieves state-of-the-art results, outperforming existing methods such as MATE in terms of both accuracy and computational efficiency. The implementation is available at: https://github.com/AliBahri94/SMART-PC.} }
Endnote
%0 Conference Paper %T SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds %A Ali Bahri %A Moslem Yazdanpanah %A Sahar Dastani %A Mehrdad Noori %A Gustavo Adolfo Vargas Hakim %A David Osowiechi %A Farzad Beizaee %A Ismail Ben Ayed %A Christian Desrosiers %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-bahri25a %I PMLR %P 2442--2457 %U https://proceedings.mlr.press/v267/bahri25a.html %V 267 %X Test-Time Training has emerged as a promising solution to address distribution shifts in 3D point cloud classification. However, existing methods often rely on computationally expensive backpropagation during adaptation, limiting their applicability in real-world, time-sensitive scenarios. In this paper, we introduce SMART-PC, a skeleton-based framework that enhances resilience to corruptions by leveraging the geometric structure of 3D point clouds. During pre-training, our method predicts skeletal representations, enabling the model to extract robust and meaningful geometric features that are less sensitive to corruptions, thereby improving adaptability to test-time distribution shifts. Unlike prior approaches, SMART-PC achieves real-time adaptation by eliminating backpropagation and updating only BatchNorm statistics, resulting in a lightweight and efficient framework capable of achieving high frame-per-second rates while maintaining superior classification performance. Extensive experiments on benchmark datasets, including ModelNet40-C, ShapeNet-C, and ScanObjectNN-C, demonstrate that SMART-PC achieves state-of-the-art results, outperforming existing methods such as MATE in terms of both accuracy and computational efficiency. The implementation is available at: https://github.com/AliBahri94/SMART-PC.
APA
Bahri, A., Yazdanpanah, M., Dastani, S., Noori, M., Vargas Hakim, G.A., Osowiechi, D., Beizaee, F., Ayed, I.B. & Desrosiers, C.. (2025). SMART-PC: Skeletal Model Adaptation for Robust Test-Time Training in Point Clouds. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:2442-2457 Available from https://proceedings.mlr.press/v267/bahri25a.html.

Related Material