EgoTac: In-the-wild Tactile Prediction from Egocentric Vision

• URL: http://arxiv.org/abs/2608.15060v1
• Date: Sat, 15 Aug 2026 06:18:13 GMT
• Title: EgoTac: In-the-wild Tactile Prediction from Egocentric Vision
• Abstract summary: We introduce EgoTac, a model that predicts rich tactile information directly from egocentric human videos.<n>EgoTac is trained on a unified corpus of over 5.7M image-tactile pairs.<n>It consistently outperforms state-of-the-art contact estimators.
• Score: 33.40360812855072
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Touch is fundamental to dexterous manipulation, yet most egocentric human data increasingly used for robot learning lacks tactile information. Directly collecting large-scale tactile data is challenging due to sensor limitations, while human video data is abundant, contact-rich, and easily scalable. This motivates a natural question: can tactile signals be inferred purely from vision? To address this, we introduce EgoTac, a generalizable model that predicts rich tactile information directly from egocentric human videos. EgoTac is trained on a unified corpus of over 5.7M image-tactile pairs, covering both continuous force measurements and binary contacts. By learning from this diverse dataset, EgoTac captures nuanced touch dynamics across varied interactions. Experiments demonstrate strong performance: in-domain prediction achieves an average force error below 0.06N. On out-of-domain contact prediction benchmarks, EgoTac consistently outperforms the state-of-the-art contact estimator. It also captures the rise and fall patterns of real tactile data and enables zero-shot predictions on unconstrained real-world videos. Scaling analyses further reveal that both data diversity and volume improve performance steadily. Overall, EgoTac provides a scalable pathway to extract tactile priors from egocentric human videos, enabling broadly applicable tactile-aware robot learning.

Related papers

• HumanScale: Egocentric Human Video Can Outperform Real-Robot Data for Embodied Pretraining [93.5192770515694]
  Embodied foundation models are expected to benefit from data scaling like large language models, but face a much tighter data bottleneck.<n>Teleoperated real-robot trajectories remain the dominant pretraining source due to their precise action supervision and embodiment alignment.<n>Embodied foundation models pretrained on egocentric data achieve a 24% lower validation loss on real-robot action prediction.
  arXiv  Detail & Related papers  (2026-06-18T17:37:34Z)
• Contrastive Action-Image Pre-training for Visuomotor Control [91.5630572541332]
  We introduce CAIP, a vision encoder that treats human hand poses from large-scale egocentric video as a proxy for end-effector actions.<n>Leveraging 32,041 hours of egocentric human video and only 88 hours of robotic manipulation data, CAIP outperforms state-of-the-art vision encoders.
  arXiv  Detail & Related papers  (2026-06-15T20:00:20Z)
• EgoTactile: Learning Grasp Pressure for Everyday Objects from Egocentric Video [82.13376072770363]
  Estimating full-hand grasp pressure from egocentric video is critical for immersive VR and robotic manipulation.<n>We introduce EgoTactile, a benchmark pairing egocentric video with full-hand pressure supervision for diverse everyday objects.<n>We also propose EgoPressureDiff, a conditional diffusion framework that adapts a large-scale pre-trained video diffusion backbone.
  arXiv  Detail & Related papers  (2026-06-08T09:20:03Z)
• TouchAnything: A Dataset and Framework for Bimanual Tactile Estimation from Egocentric Video [20.373348802426143]
  EgoTouch is a large-scale egocentric dataset with dense tactile supervision for hand-object interaction.<n>We introduce TouchAnything, a vision-to-touch prediction framework that uses the egocentric view as the primary input.<n>We will publicly release the dataset, code, and benchmark.
  arXiv  Detail & Related papers  (2026-05-13T06:54:36Z)
• OPENTOUCH: Bringing Full-Hand Touch to Real-World Interaction [93.88239833545623]
  We present OpenTouch, the first in-the-wild egocentric full-hand tactile dataset.<n>We show that tactile signals provide a compact yet powerful cue for grasp understanding.<n>We aim to advance multimodal egocentric perception, embodied learning, and contact-rich robotic manipulation.
  arXiv  Detail & Related papers  (2025-12-18T18:18:17Z)
• Grasp Like Humans: Learning Generalizable Multi-Fingered Grasping from Human Proprioceptive Sensorimotor Integration [26.351720551267846]
  Tactile and kinesthetic perceptions are crucial for human dexterous manipulation, enabling reliable grasping of objects via sensorimotor integration.<n>We propose a novel glove-mediated tactile-kinematic perception-prediction framework for grasp skill transfer from human intuitive and natural operation to robotic execution based on imitation learning.
  arXiv  Detail & Related papers  (2025-09-10T07:44:12Z)
• EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video [7.1221123957033905]
  EgoDex is the largest and most diverse dataset of dexterous human manipulation to date.<n>It has 829 hours of egocentric video with paired 3D hand and finger tracking data collected at the time of recording.<n>The dataset covers a wide range of diverse manipulation behaviors with everyday household objects in 194 different tabletop tasks.
  arXiv  Detail & Related papers  (2025-05-16T21:34:47Z)
• Digitizing Touch with an Artificial Multimodal Fingertip [51.7029315337739]
  Humans and robots both benefit from using touch to perceive and interact with the surrounding environment. Here, we describe several conceptual and technological innovations to improve the digitization of touch. These advances are embodied in an artificial finger-shaped sensor with advanced sensing capabilities.
  arXiv  Detail & Related papers  (2024-11-04T18:38:50Z)
• EgoMimic: Scaling Imitation Learning via Egocentric Video [22.902881956495765]
  We present EgoMimic, a full-stack framework which scales manipulation via human embodiment data. EgoMimic achieves this through: (1) a system to capture human embodiment data using the ergonomic Project Aria glasses, (2) a low-cost bimanual manipulator that minimizes the kinematic gap to human data, and (4) an imitation learning architecture that co-trains on human and robot data.
  arXiv  Detail & Related papers  (2024-10-31T17:59:55Z)
• Combining Vision and Tactile Sensation for Video Prediction [0.0]
  We investigate the impact of integrating tactile feedback into video prediction models for physical robot interactions. We introduce two new datasets of robot pushing that use a magnetic-based tactile sensor for unsupervised learning. Our results demonstrate that incorporating tactile feedback into video prediction models improves scene prediction accuracy and enhances the agent's perception of physical interactions.
  arXiv  Detail & Related papers  (2023-04-21T18:02:15Z)
• Ego-Exo: Transferring Visual Representations from Third-person to First-person Videos [92.38049744463149]
  We introduce an approach for pre-training egocentric video models using large-scale third-person video datasets. Our idea is to discover latent signals in third-person video that are predictive of key egocentric-specific properties. Our experiments show that our Ego-Exo framework can be seamlessly integrated into standard video models.
  arXiv  Detail & Related papers  (2021-04-16T06:10:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.