analiza aiqbench
This research introduces LESS, a novel machine learning framework for tactile imaging that utilizes local scene representations to overcome the generalization limits of previous global models. By employing recurrent encoders with local receptive fields, the system achieves robust reconstruction of soft object structures even when trained on limited data and adapted to hand-held sensing scenarios. This advancement is significant for robotics and medical diagnostics where efficient, adaptable touch-based perception is critical.
More with LESS -- Local Scene Representations for Tactile Imaging
arXiv:2606.14344v1 Announce Type: new Abstract: Tactile imaging seeks to reconstruct the internal structure of soft objects through touch sensing, with applications in medical diagnosis and robotic manipulation. Recent self-supervised learning approaches have shown promising results, but rely on global, unstructured representations and robot-controlled sensing, limiting generalization and practical use. We propose Local Encoder for Spatial Sensing (LESS), an object-centric tactile representation that exploits the local nature of touch. The tactile scene is modeled as a grid of recurrent encoders with local receptive fields, whose states are fused to reconstruct 2D or 3D images of internal structure. This compositional design enables strong generalization: models trained on single-inclusion phantoms accurately image objects with multiple inclusions and varying sizes. The local structure further supports spatial uncertainty estimation. In addition, we enable hand-held tactile imaging via external pose tracking and human-like palpation data, and extend tactile imaging to full 3D reconstruction.
Abstract:Tactile imaging seeks to reconstruct the internal structure of soft objects through touch sensing, with applications in medical diagnosis and robotic manipulation. Recent self-supervised learning approaches have shown promising results, but rely on global, unstructured representations and robot-controlled sensing, limiting generalization and practical use. We propose Local Encoder for Spatial Sensing (LESS), an object-centric tactile representation that exploits the local nature of touch. The tactile scene is modeled as a grid of recurrent encoders with local receptive fields, whose states are fused to reconstruct 2D or 3D images of internal structure. This compositional design enables strong generalization: models trained on single-inclusion phantoms accurately image objects with multiple inclusions and varying sizes. The local structure further supports spatial uncertainty estimation. In addition, we enable hand-held tactile imaging via external pose tracking and human-like palpation data, and extend tactile imaging to full 3D reconstruction.
Machine Learning (cs.LG)
arXiv:2606.14344 [cs.LG]
(or arXiv:2606.14344v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2606.14344 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 236 cuvinte
Subiecte conexe
$\mathcal{N}_0$-Foundation: Towards the Age of Tactile Intelligence
arXiv · 2 sept. 2026
HT-Bench: Benchmarking and Learning Dexterous Full-Hand Tactile Representations with Egocentric Vision
arXiv · 21 aug. 2026
Factorized Tactile Representation and Control for Sim-to-Real Manipulation
arXiv · 8 oct. 2026
Learning Tactile-Aware Hand Representations from Human Hand-Object Interaction Data
arXiv · 21 oct. 2026