SaccadeCam: Adaptive Visual Attention for Monocular Depth Sensing

TL;DR

SaccadeCam introduces an adaptive, self-supervised visual attention mechanism inspired by animal saccades to improve monocular depth sensing, demonstrating promising results in both simulation and hardware prototype.

Contribution

The paper presents a novel self-supervised network for adaptive resolution in monocular depth estimation, inspired by animal eye saccades, with initial hardware prototype results.

Findings

Effective adaptive resolution distribution improves depth estimation.

End-to-end learning enhances monocular depth sensing.

Preliminary hardware results validate the approach.

Abstract

Most monocular depth sensing methods use conventionally captured images that are created without considering scene content. In contrast, animal eyes have fast mechanical motions, called saccades, that control how the scene is imaged by the fovea, where resolution is highest. In this paper, we present the SaccadeCam framework for adaptively distributing resolution onto regions of interest in the scene. Our algorithm for adaptive resolution is a self-supervised network and we demonstrate results for end-to-end learning for monocular depth estimation. We also show preliminary results with a real SaccadeCam hardware prototype.