activity
20172021
most citedEnvironment Predictive Coding for Embodied Agents

5 citations · 17 across the 5 of their papers we have counts for

collaborators
Showing cs.CVShow all

8 papers · 1 filter

cs.CV20213 cited

Shaping embodied agent behavior with activity-context priors from egocentric video

Tushar Nagarajan, Kristen Grauman

Complex physical tasks entail a sequence of object interactions, each with its own preconditions -- which can be difficult for robotic agents to learn efficiently solely through th…

cs.CV20215 cited

Ego-Exo: Transferring Visual Representations from Third-person to First-person Videos

Yanghao Li, Tushar Nagarajan, Bo Xiong +1

We introduce an approach for pre-training egocentric video models using large-scale third-person video datasets. Learning from purely egocentric data is limited by low dataset scal…

cs.CV20215 cited

Environment Predictive Coding for Embodied Agents

Santhosh K. Ramakrishnan, Tushar Nagarajan, Ziad Al-Halah +1

We introduce environment predictive coding, a self-supervised approach to learn environment-level representations for embodied agents. In contrast to prior work on self-supervised…

cs.CV2020

Learning Affordance Landscapes for Interaction Exploration in 3D Environments

Tushar Nagarajan, Kristen Grauman

Embodied agents operating in human spaces must be able to master how their environment works: what objects can the agent use, and how can it use them? We introduce a reinforcement…

cs.CV2020

EGO-TOPO: Environment Affordances from Egocentric Video

Tushar Nagarajan, Yanghao Li, Christoph Feichtenhofer +1

First-person video naturally brings the use of a physical environment to the forefront, since it shows the camera wearer interacting fluidly in a space based on his intentions. How…

cs.CV2019

Grounded Human-Object Interaction Hotspots from Video (Extended Abstract)

Tushar Nagarajan, Christoph Feichtenhofer, Kristen Grauman

Learning how to interact with objects is an important step towards embodied visual intelligence, but existing techniques suffer from heavy supervision or sensing requirements. We p…