Research map / Vision-language and multimodal / Robot action and navigation
AnyTouch 2: General Optical Tactile Representation Learning For Dynamic Tactile Perception
Ruoxuan Feng, Yuxuan Zhou, Siyu Mei, Dongzhan Zhou, Pengwei Wang, Shaowei Cui, Bin Fang, Guocai Yao, Di Hu
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 frame-difference reconstruction (ablation) | AnyTouch 2 · ToucHD Bench Force (DIGIT) | RMSE (mN): 1009.44 -> 894.32 | Improved |
| add action matching (ablation) | AnyTouch 2 · Sparsh Slip (DIGIT) | F1 score: 84.42 -> 86.66 | Improved |
| add force and delta-force prediction (ablation) | AnyTouch 2 · ToucHD Bench Force (DIGIT) | RMSE (mN): 1646.95 -> 894.32 | Improved |
| add force prediction (ablation) | AnyTouch 2 · Sparsh Force (DIGIT) | RMSE (mN): 770.44 -> 624.26 | Improved |
| add multi-modal alignment (ablation) | AnyTouch 2 · TAG (GelSight) | accuracy: 63.84 -> 76.97 | Improved |
| add multi-modal alignment (ablation) | AnyTouch 2 · ToucHD Bench Force (GelSight Mini) | RMSE (mN): 972.37 -> 1051.03 (RMSE bests) | Hurt |
| add ToucHD (Sim) subset (ablation) | AnyTouch 2 · Sparsh Slip delta force (GelSight Mini) | RMSE (mN): 108.25 -> 80.83 | Improved |
| add ToucHD (Mani) subset (ablation) | AnyTouch 2 · ToucHD Bench Force (DIGIT) | RMSE (mN): 1118.49 -> 894.32 | Improved |
| add ToucHD (Force) subset (ablation) | AnyTouch 2 · ToucHD Bench Force (GelSight Mini) | RMSE (mN): 2424.68 -> 1051.03 | Improved |
| add the full ToucHD dataset (ablation) | AnyTouch 2 · ToucHD Bench Force (DIGIT) | RMSE (mN): 2448.89 -> 894.32 | Improved |
| add the full ToucHD dataset (ablation) | AnyTouch 2 · TAG (GelSight) | accuracy: 68.92 -> 76.97 | Improved |
| train MAE (Sparsh) with additional data including ToucHD | MAE (Sparsh) · Sparsh Force (DIGIT) | RMSE (mN): 821.26 -> 741.67 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- AnyTouch 1 — AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors (ICLR 2025) · better on 2 of 3 shared results
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
- AnyTouch: Learning Unified Static-Dynamic Representation across Multiple Visuo-tactile Sensors (ICLR 2025 · similarity 0.93)
- Touch in the Wild: Learning Fine-Grained Manipulation with a Portable Visuo-Tactile Gripper (NeurIPS 2025 · similarity 0.91)
- Sensor-Invariant Tactile Representation (ICLR 2025 · similarity 0.90)
- PhysTacGen: Physics-Aware Visual-Tactile Sensor Image Generation (NeurIPS 2026 · similarity 0.90)
- EgoTac: In-the-wild Tactile Prediction from Egocentric Vision (NeurIPS 2026 · similarity 0.90)
- Touch-R1: Reinforcing Touch Reasoning in MLLMs (NeurIPS 2026 · similarity 0.90)
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.
