Learning Monocular Depth from Focus with Event Focal Stack

TL;DR

This paper introduces EDFF, a novel neural network that leverages event camera data to improve monocular depth estimation from focal stacks, overcoming traditional camera sampling limitations.

Contribution

The study proposes a new event-based focal stack depth estimation method with specialized encoding, attention modules, and multi-level fusion, outperforming existing approaches.

Findings

Outperforms state-of-the-art methods in depth accuracy

Utilizes event camera data for enhanced focus cue extraction

Demonstrates robustness in low-latency focus measurement

Abstract

Depth from Focus estimates depth by determining the moment of maximum focus from multiple shots at different focal distances, i.e. the Focal Stack. However, the limited sampling rate of conventional optical cameras makes it difficult to obtain sufficient focus cues during the focal sweep. Inspired by biological vision, the event camera records intensity changes over time in extremely low latency, which provides more temporal information for focus time acquisition. In this study, we propose the EDFF Network to estimate sparse depth from the Event Focal Stack. Specifically, we utilize the event voxel grid to encode intensity change information and project event time surface into the depth domain to preserve per-pixel focal distance information. A Focal-Distance-guided Cross-Modal Attention Module is presented to fuse the information mentioned above. Additionally, we propose a Multi-level Depth Fusion Block designed to integrate results from each level of a UNet-like architecture and produce the final output. Extensive experiments validate that our method outperforms existing state-of-the-art approaches.