Problem
Research questions and friction points this paper is trying to address.
Monocular Depth Estimation
Event Camera
Depth Regression
Ill-posed Problem
Depth Hypothesis Guided Iterative Refinement for Event-Image Monocular Depth Estimation
🤖 AI Summary
This study addresses the ill-posed and highly nonlinear challenges of direct regression in monocular depth estimation from event cameras by proposing HypoDepth, the first event-image iterative refinement framework. HypoDepth introduces discrete depth hypothesis volumes and 3D cost volume construction to reformulate continuous regression as a constrained search problem. It achieves efficient multi-resolution refinement from global to local scales via multi-scale correlation guidance, further enhancing accuracy through a lightweight residual module. Experimental results demonstrate that the proposed method attains state-of-the-art performance on the DSEC and MVSEC datasets while exhibiting strong zero-shot generalization capabilities. Moreover, the model remains computationally lightweight, enabling real-time inference.
📝 Abstract
Event cameras hold excellent dynamic properties, showing great potential for monocular depth estimation (MDE). However, existing methods mainly improve performance by optimizing contextual features, but still struggle with the ill-posed and nonlinear nature of direct full-depth regression. In this paper, we propose HypoDepth, the first event-image monocular depth iterative refinement framework. By introducing a discrete Depth Hypothesis Volume (DHV), we transform the depth regression problem into a constrained depth search task. Specifically, we construct a 3D cost volume between the DHV features and contextual features and perform a multi-scale correlation search to guide stable residual optimization. This lightweight cost volume enables efficient global-to-local refinement across multi-resolution. Our method outperforms existing approaches on DSEC and MVSEC with state-of-the-art results and strong zero-shot generalization. Meanwhile, our tiny model achieves an excellent balance between accuracy and efficiency, enabling real-time performance on resource-limited devices.
Innovation
Methods, ideas, or system contributions that make the work stand out.
Event-Image Monocular Depth Estimation
Iterative Refinement
Depth Hypothesis Volume
3D Cost Volume
Multi-scale Correlation Search
🔎 Similar Papers
No similar papers found.
