Research map / Point clouds / Point cloud classification
UST-SSM: Unified Spatio-Temporal State Space Models for Point Cloud Video Modeling
Peiming Li, Ziyi Wang, Yulin Yuan, Hong Liu, Xiangming Meng, Junsong Yuan, Mengyuan Liu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add selective SSM (ablation) | P4Transformer · MSR-Action3D | accuracy (%): 90.94 -> 92.33 | Improved |
| add STSA to ST-SSM (ablation) | P4Transformer + SSM · MSR-Action3D | accuracy (%): 92.33 -> 93.03 | Improved |
| add STSS (ablation) | P4Transformer + SSM + STSA · MSR-Action3D | accuracy (%): 93.03 -> 94.43 | Improved |
| add TIS temporal interaction sampling (ablation) | UST-SSM · MSR-Action3D | accuracy (%): 94.43 -> 94.77 | Improved |
| use spatio-temporal selection with Hilbert curve instead of cross-temporal scanning (ablation) | UST-SSM · MSR-Action3D | accuracy (%): 89.20 -> 94.77 | Improved |
| use spatio-temporal selection instead of temporally sequential Hilbert scanning (ablation) | UST-SSM · MSR-Action3D | accuracy (%): 94.08 -> 94.77 | Improved |
| use TIS (strides 1+2) instead of single stride 1 (ablation) | UST-SSM · MSR-Action3D | memory (GB): 3.079 -> 1.759 | Improved |
| add STSS (ablation) | UST-SSM without STSA · NTU RGB+D | accuracy (%): 91.4 -> 92.1 | Improved |
| add STSS and STSA (ablation) | UST-SSM · NTU RGB+D | accuracy (%): 90.8 -> 92.1 | Improved |
Later papers that compared with this method
1 later accepted paper reports UST-SSM as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- DAP: Doppler-aware Point Network for Heterogeneous mmWave Action Recognition (ECCV 2026 · better on 2 of 2 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- MAMBA4D — Mamba4D: Efficient 4D Point Cloud Video Understanding with Disentangled Spatial-Temporal State Space Models (CVPR 2025) · better on 3 of 3 shared results
- PSTNet — PSTNet: Point Spatio-Temporal Convolution on Point Cloud Sequences (ICLR 2021) · better on 3 of 3 shared results
- PointCMP — PointCMP: Contrastive Mask Prediction for Self-Supervised Learning on Point Cloud Videos (CVPR 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Point Spatio-Temporal Transformer Networks for Point Cloud Video Modeling (TPAMI 2022 · similarity 0.90)
- Adapting Pre-trained 3D Models for Point Cloud Video Understanding via Cross-frame Spatio-temporal Perception (CVPR 2025 · similarity 0.89)
- 3DInAction: Understanding Human Actions in 3D Point Clouds (CVPR 2024 · similarity 0.88)
- PointMamba: A Simple State Space Model for Point Cloud Analysis (NeurIPS 2024 · similarity 0.86)
- M3SOT: Multi-Frame, Multi-Field, Multi-Space 3D Single Object Tracking (AAAI 2024 · similarity 0.86)
- State Space Model Meets Transformer: A New Paradigm for 3D Object Detection (ICLR 2025 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
