Work

HybridDepth: Robust Metric Depth Fusion by Leveraging Depth from Focus and Single-Image Priors

Year: 2025

Type: conference-paper

Abstract: We propose Hybriddepth, a robust depth estimation pipeline that addresses key challenges in depth estimation, including scale ambiguity, hardware heterogene-ity, and generalizability. Hybriddepth leverages focal stack, data conveniently accessible in common mobile de-vices, to produce accurate metric depth maps. By incorpo-rating depth priors afforded by recent advances in single-image depth estimation, our model achieves a higher level of structural detail compared to existing methods. We test our pipeline as an end-to-end system, with a newly developed mobile client to capture focal stacks, which are then sent to a GPU-powered server for depth estimation. Comprehensive quantitative and qualitative analyses demonstrate that Hybriddepth outperforms state-of-the-art (SOTA) models on common datasets such as DDFF12 and NYU Depth V2. Hybriddepth also shows strong zero-shot generalization. When trained on NYU Depth V2, Hybriddepth surpasses SOTA models in zero-shot per-formance on ARKitScenes and delivers more structurally accurate depth maps on Mobile Depth. The code is avail-able at https://github.com/cake-labIHybridDepth/.

Source: IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

DOI: 10.1109/wacv61041.2025.00104

Authors: Ashkan Ganj, Hang Su, Tian Guo

Institutions: Worcester Polytechnic Institute, Nvidia (United Kingdom), Nvidia (United States)

Language: English

Topic: Advanced Vision and Imaging

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: National Science Foundation

Awards: Collaborative Research: SaTC: CORE: Small: Towards the Security of Immersive Multimedia Systems, CAREER: Toward a Specialized Edge for Augmented Reality, Collaborative Research: CIRC: Planning-C: ExpAR: Scalable and Controllable AR Experimentation