Balancing Marker and Markerless Modes in Vision-Based Tactile Sensors with a Translucent Skin
Oluwatimilehin Tijani, Zhuo Chen, Jiankang Deng, SHAN LUO
AI summary
Problem
Vision-based tactile sensors traditionally force a choice between opaque markers for accurate tracking or markerless skins for surface detail, as existing hybrid solutions require complex hardware or heavy computation.
Approach
The authors engineered an elastomer skin with integrated translucent, tinted markers that remain visible for displacement tracking while preserving optical clarity for fine surface geometry capture.
Key results
- 99.17% object and 93.51% texture classification accuracy
- 97% point retention for tangential displacement tracking
- 66% reduction in total force prediction error
- Dual-mode sensing without hardware or software switching
Why it matters
Provides roboticists with a low-cost, plug-and-play tactile skin that simultaneously captures geometry, texture, and force, streamlining multi-modal perception for manipulation tasks.
Abstract
Vision-based tactile sensors (VBTS) face an in- herent trade-off in tactile skin design. Opaque ink markers enable accurate force and tangential displacement estimation but occlude geometric features essential for object and texture classification. Conversely, markerless skins preserve surface details yet provide limited capability for tangential motion estimation. Existing approaches, including UV illumination and learning-based virtual marker transfer, increase hardware complexity or computational cost. We present a novel tactile skin with translucent, tinted markers balancing the modes of marker and markerless for VBTS. This design enables concurrent tangential displacement tracking, force estimation, and preservation of surface geometry. It integrates directly with the GelSight sensor family, requiring no additional hardware and minimal software modification. Experimental evaluation demonstrates that translucent skin improves overall sensing performance relative to both opaque-marker and markerless configurations. It achieves 99.17% accuracy in object classifi- cation, 93.51% in texture classification, 97% point retention in tangential displacement tracking, and a 66% reduction in total force error. These results indicate that translucent skin substantially mitigate the traditional trade-off between marker- based and markerless tactile sensing, thereby expanding the applicability of multi-modal VBTS in tactile robotics.