From Time to Depth: A Unified Framework for 3D Vision and Action

Read the full article See related articles

Listed in

This article is not in any list yet, why not save it to one of your lists.
Log in to save this article

Abstract

Understanding how organisms perceive and interact with three-dimensional (3D) structure in a 3D world based on two-dimensional (2D) optic images is a fundamental question in vision science. Traditional approaches have often relied on a perception–cognition–action sequence, wherein spatial representations are first constructed from indirect depth cues and subsequently used to plan actions. In this review, an alternative framework inspired by ecological theory is proposed: unveiling the third dimension through the explicit integration of a fourth dimension—time. Time-dimensioned optical variables provide direct, continuously updated information that supports both the recovery of 3D structure and the prospective control of action. Rather than relying on internal spatial maps and feedforward predictions, organisms can achieve real-time, dynamic coupling with their environment by perceiving time-dimensioned optic variables (aka tau models). Drawing on evidence from perceptual psychology, neuroscience, and sensorimotor control, it is highlighted how tau and tau related variables provide robust information for interacting with objects in depth. This perspective not only aligns with the continuous, dynamic nature of natural behavior but also offers a unified framework of perception and action that emphasizes real-time coupling with the environment

Article activity feed