Sparsh: Self-supervised touch representations for vision-based tactile sensing
Reviewed by Pithpith:H7T5QBREopen to challenge →
read the original abstract
In this work, we introduce general purpose touch representations for the increasingly accessible class of vision-based tactile sensors. Such sensors have led to many recent advances in robot manipulation as they markedly complement vision, yet solutions today often rely on task and sensor specific handcrafted perception models. Collecting real data at scale with task centric ground truth labels, like contact forces and slip, is a challenge further compounded by sensors of various form factor differing in aspects like lighting and gel markings. To tackle this we turn to self-supervised learning (SSL) that has demonstrated remarkable performance in computer vision. We present Sparsh, a family of SSL models that can support various vision-based tactile sensors, alleviating the need for custom labels through pre-training on 460k+ tactile images with masking and self-distillation in pixel and latent spaces. We also build TacBench, to facilitate standardized benchmarking across sensors and models, comprising of six tasks ranging from comprehending tactile properties to enabling physical perception and manipulation planning. In evaluations, we find that SSL pre-training for touch representation outperforms task and sensor-specific end-to-end training by 95.1% on average over TacBench, and Sparsh (DINO) and Sparsh (IJEPA) are the most competitive, indicating the merits of learning in latent space for tactile images. Project page: https://sparsh-ssl.github.io/
This paper has not been read by Pith yet.
Forward citations
Cited by 15 Pith papers
-
TacVerse: A Multi-Sensor Dataset and Benchmark for Cross-Sensor Vision-Based Tactile Perception
TacVerse is a new multi-sensor tactile dataset with 106,800 images from seven VBTS designs that benchmarks within-sensor performance, zero-shot cross-sensor transfer, and few-shot adaptation on shape, grating, and for...
-
FTP-1: A Generalist Foundation Tactile Policy Across Tactile Sensors for Contact-Rich Manipulation
FTP-1 is the first foundation tactile policy pretrained on ~3000 hours of data from 26 sources across 21 sensors that improves performance on seen setups by 17.2% and transfers to unseen sensors with 31% success rate gain.
-
Touch-R1: Reinforcing Touch Reasoning in MLLMs
Touch-R1 applies GRPO reinforcement learning on a new 1M tactile dataset and benchmark to train a Qwen2.5-VL-7B model that outperforms baselines on tactile perception and visual-tactile conflict tasks.
-
AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models
AT-VLA proposes adaptive tactile injection and a dual-stream tactile reaction mechanism to enhance VLA models for contact-rich robotic manipulation with real-time responses.
-
TactX: Learning Shared Tactile Representations Across Diverse Sensors
TactX learns a shared latent representation across three tactile sensor modalities via joint training on paired contacts, enabling zero-shot policy transfer and higher success on pick-and-place, insertion, wiping, and...
-
Tactile-WAM: Touch-Aware World Action Model with Tactile Asymmetric Attention
Tactile-WAM with TAAM improves mean success rate by 38.9% overall and 86% on contact-rich tasks on ManiFeel by using VideoClean mask and touch-aware bias to prevent tactile pollution in world action models.
-
Imitation from Heterogeneous Demonstrations using Grounded Latent-Action World Models
GLAM learns a shared latent action space grounded in consistent future observation prediction across heterogeneous data sources to train improved behavioral cloning policies for robot manipulation tasks.
-
Transferring Contact, Not Just Motion: Compliant Grasping Across Dexterous Hands
A cross-embodiment force-position interface with system-identified torque calibration enables a flow-matching policy to perform transferable compliant grasping on heterogeneous dexterous hands.
-
AT-VLA: Adaptive Tactile Injection for Enhanced Feedback Reaction in Vision-Language-Action Models
AT-VLA introduces adaptive tactile injection and a dual-stream tactile reaction mechanism to integrate real-time tactile feedback into pretrained VLA models for contact-rich robotic manipulation.
-
Learning to Feel the Future: DreamTacVLA for Contact-Rich Manipulation
DreamTacVLA grounds VLA models in contact physics by aligning multi-scale vision-tactile inputs and predicting future tactile states, reaching up to 95% success on contact-rich tasks.
-
ViTacFormer: Learning Cross-Modal Representation for Visuo-Tactile Dexterous Manipulation
ViTacFormer learns a cross-modal visuo-tactile latent space with autoregressive tactile prediction and an easy-to-hard curriculum, then uses the representation for imitation learning that yields ~50% higher success an...
-
TacEvo: Self-Evolving Architecture Discovery for Robotic Tactile Perception via LLM-Driven Quality-Diversity Search
TacEvo is an LLM-driven self-evolving search method that discovers neural architectures for robotic tactile force regression and grating classification, reporting fitness gains of 56.1% and 96.1% over 20 generations.
-
Seeing Touch from Motion: A Unified Modality-Aware Visuo-Tactile Policy with Tactile Motion Correlation
A visuo-tactile policy learning method that exploits tactile motion correlation for contact state distinction and Mixture-of-Transformers for cross-modal fusion.
-
TouchThinker: Scaling Tactile Commonsense Reasoning to the Open World with Large-scale Data and Action-aware Representation
TouchThinker introduces a 1M-scale multi-source tactile dataset and action-aware modeling to scale commonsense reasoning from tactile observations, reporting competitive performance on new and existing benchmarks.
-
Characterizing the Resilience and Sensitivity of Polyurethane Vision-Based Tactile Sensors
Polyurethane vision-based tactile sensors are more resilient to normal loading, shear, and abrasion than silicone ones, extending the usable force range at the cost of low-force sensitivity.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.