Enhancing Robot Learning through Learned Human-Attention Feature Maps

TL;DR

This paper introduces a method to improve robot learning by modeling human attention and integrating it as auxiliary feature maps, enhancing robustness and learning speed in visual tasks.

Contribution

The paper presents a novel approach to emulate human attention and incorporate it into robot learning, improving performance in object detection and imitation learning tasks.

Findings

Improved robustness to out-of-distribution samples

Faster learning in low-data regimes

Effective integration of human attention into robot perception

Abstract

Robust and efficient learning remains a challenging problem in robotics, in particular with complex visual inputs. Inspired by human attention mechanism, with which we quickly process complex visual scenes and react to changes in the environment, we think that embedding auxiliary information about focus point into robot learning would enhance efficiency and robustness of the learning process. In this paper, we propose a novel approach to model and emulate the human attention with an approximate prediction model. We then leverage this output and feed it as a structured auxiliary feature map into downstream learning tasks. We validate this idea by learning a prediction model from human-gaze recordings of manual driving in the real world. We test our approach on two learning tasks - object detection and imitation learning. Our experiments demonstrate that the inclusion of predicted human attention leads to improved robustness of the trained models to out-of-distribution samples and faster learning in low-data regime settings. Our work highlights the potential of incorporating structured auxiliary information in representation learning for robotics and opens up new avenues for research in this direction. All code and data are available online.