Does computer vision matter for action?

Abstract

Computer vision produces representations of scene content. Much computer vision research is predicated on the assumption that these intermediate representations are useful for action. Recent work at the intersection of machine learning and robotics calls this assumption into question by training sensorimotor systems directly for the task at hand, from pixels to actions, with no explicit intermediate representations. Thus the central question of our work: Does computer vision matter for action? We probe this question and its offshoots via immersive simulation, which allows us to conduct controlled reproducible experiments at scale. We instrument immersive three-dimensional environments to simulate challenges such as urban driving, off-road trail traversal, and battle. Our main finding is that computer vision does matter. Models equipped with intermediate representations train faster, achieve higher task performance, and generalize better to previously unseen environments. A video that summarizes the work and illustrates the results can be found at https://youtu.be/4MfWa2yZ0Jc

arXiv commentPublished in Science Robotics, 4(30), May 2019

Cited by in corpus (7)

• Learning High-Speed Flight in the Wild
• A Survey of End-to-End Driving: Architectures and Training Methods
• Navigation-Oriented Scene Understanding for Robotic Autonomy: Learning to Segment Driveability in Egocentric Images
• Self-Supervised Deep Pose Corrections for Robust Visual Odometry
• MUSES: The Multi-Sensor Semantic Perception Dataset for Driving under Uncertainty
• Domain Adaptation Through Task Distillation
• Investigation of Factorized Optical Flows as Mid-Level Representations