Antonio Torralba

About

Antonio Torralba is a prominent researcher whose work spans computer vision, robotics, tactile sensing, and human-machine interaction — fields where his contributions have helped redefine how machines perceive and interact with the physical world. Perhaps most recognized for his pioneering work on tactile sensing, Torralba co-developed scalable tactile glove systems capable of learning the signatures of human grasping (1,118 citations), with subsequent advances in conformal tactile textiles (375 citations) and digitally embroidered smart gloves pushing this technology toward real-world wearable applications. His research consistently bridges the gap between touch and vision, as demonstrated by his influential work on cross-modal prediction connecting these two sensory domains (126 citations). In robotics and scene understanding, Torralba has contributed meaningfully to 3D semantic representations through ConceptGraphs (178 citations) and integrated perception systems like SegICP (151 citations), enabling robots to better reason about and manipulate their environments. His explorations of generative video modeling (175 citations) and particle-based physical simulation (107 citations) further illustrate the remarkable breadth of his research portfolio. With thousands of cumulative citations across diverse domains, Torralba's work remains essential reading for researchers working at the intersection of perception, embodied AI, and intelligent systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Learning the signatures of the human grasp using a scalable tactile glove

   1,118 citations · 2019
2. 2
   Learning human–environment interactions using conformal tactile textiles

   375 citations · 2021
3. 3
   ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning

   178 citations · 2024
4. 4
   Generating the Future with Adversarial Transformers

   175 citations · 2017
5. 5
   SegICP: Integrated deep semantic segmentation and pose estimation

   151 citations · 2017
6. 6
   Connecting Touch and Vision via Cross-Modal Prediction

   126 citations · 2019
7. 7
   Adaptive tactile interaction transfer via digitally embroidered smart gloves

   118 citations · 2024
8. 8
   Self-powered sensing systems with learning capability

   114 citations · 2022
9. 9
   Learning Particle Dynamics for Manipulating Rigid Bodies, Deformable Objects, and Fluids

   107 citations · 2018
10. 10
    Correcting Robot Plans with Natural Language Feedback

    67 citations · 2022

Key Collaborators

Contact & Links