Visual Perspective Taking for Opponent Behavior Modeling

TL;DR

This paper introduces an end-to-end visual prediction framework enabling robots to perform visual perspective taking and behavior modeling, demonstrated through a long-term prediction in a visual hide-and-seek game, advancing social interaction capabilities.

Contribution

The work presents a novel long-term visual prediction model that extrapolates multiple future frames, integrating perspective taking and behavior modeling for robots in social scenarios.

Findings

Predicts 25 seconds into the future, 175% beyond training horizon.

Enables robots to infer others' views and plans in social games.

Improves robots' social interaction skills in multi-agent environments.

Abstract

In order to engage in complex social interaction, humans learn at a young age to infer what others see and cannot see from a different point-of-view, and learn to predict others' plans and behaviors. These abilities have been mostly lacking in robots, sometimes making them appear awkward and socially inept. Here we propose an end-to-end long-term visual prediction framework for robots to begin to acquire both these critical cognitive skills, known as Visual Perspective Taking (VPT) and Theory of Behavior (TOB). We demonstrate our approach in the context of visual hide-and-seek - a game that represents a cognitive milestone in human development. Unlike traditional visual predictive model that generates new frames from immediate past frames, our agent can directly predict to multiple future timestamps (25s), extrapolating by 175% beyond the training horizon. We suggest that visual behavior modeling and perspective taking skills will play a critical role in the ability of physical robots to fully integrate into real-world multi-agent activities. Our website is at http://www.cs.columbia.edu/~bchen/vpttob/.