Research map / Vision-language and multimodal / Robot action and navigation
Seeing Touch from Motion: A Unified Modality-Aware Visuo-Tactile Policy with Tactile Motion Correlation
Shengqi Xu, Yang Liu, Guojin Zhong, Fanjie Wang, Hu Luo, Hanyu Zhou, WeiYao Zhang, Ziyi Ye, Zuxuan Wu, Yu-Gang Jiang
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 tactile motion correlation (TMC) representation (ablation) | ViTacMotor base · Whiteboard Erasing | success rate (%): 46.7 -> 73.3 | Improved |
| add TMC representation (ablation) | ViTacMotor base · Tube Collection | success rate (%): 53.3 -> 66.7 | Improved |
| add MoT-based fusion (ablation) | ViTacMotor base · Whiteboard Erasing | success rate (%): 46.7 -> 66.7 | Improved |
| add MoT-based fusion (ablation) | ViTacMotor base · Tube Collection | success rate (%): 53.3 -> 60.0 | Improved |
| combine TMC and MoT fusion (ablation) | ViTacMotor · Whiteboard Erasing | success rate (%): 73.3 -> 86.7 | Improved |
| use TMC instead of cumulative motion (ablation) | ACT · Whiteboard Erasing | success rate (%): 53.3 -> 73.3 | Improved |
| use TMC instead of raw tactile image (ablation) | DP · Whiteboard Erasing | success rate (%): 53.3 -> 66.7 | Improved |
| cumulative motion instead of raw image (ablation) | ACT and DP · Whiteboard Erasing | success rate (%): 60.0 -> 53.3 (ACT); 53.3 -> 40.0 (DP) | Hurt |
| MoT fusion instead of concatenation (ablation) | ViTacMotor · Whiteboard Erasing and Tube Collection (avg) | success rate (%): 70.0 -> 80.0 | Improved |
| MoT fusion instead of cross attention (ablation) | ViTacMotor · Whiteboard Erasing and Tube Collection (avg) | success rate (%): 66.7 -> 80.0 | Improved |
| add raw tactile image to ACT | ACT · Tube Collection | success rate (%): 53.3 -> 60.0 | Improved |
| ViTacMotor vs TactileACT | ViTacMotor · Lightbulb Insertion (whole task) | success rate (%): 40.0 vs 40.0 | No clear effect |
Closest papers by meaning
- AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors (ICLR 2025 · similarity 0.90)
- Touch in the Wild: Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper (NeurIPS 2025 · similarity 0.90)
- EgoTac: In-the-wild Tactile Prediction from Egocentric Vision (NeurIPS 2026 · similarity 0.89)
- PhysTacGen: Physics-Aware Visual-Tactile Sensor Image Generation (NeurIPS 2026 · similarity 0.89)
- DIFFTACTILE: A Physics-based Differentiable Tactile Simulator for Contact-rich Robotic Manipulation (ICLR 2024 · similarity 0.89)
- Universal Visuo-Tactile Video Understanding for Embodied Interaction (NeurIPS 2025 · similarity 0.88)
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.
