SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Pham, Duc-Hai, Do, Tung, Nguyen, Phong, Hua, Binh-Son, Nguyen, Khoi, Nguyen, Rang
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913588365492224
author Pham, Duc-Hai
Do, Tung
Nguyen, Phong
Hua, Binh-Son
Nguyen, Khoi
Nguyen, Rang
author_facet Pham, Duc-Hai
Do, Tung
Nguyen, Phong
Hua, Binh-Son
Nguyen, Khoi
Nguyen, Rang
contents We propose SharpDepth, a novel approach to monocular metric depth estimation that combines the metric accuracy of discriminative depth estimation methods (e.g., Metric3D, UniDepth) with the fine-grained boundary sharpness typically achieved by generative methods (e.g., Marigold, Lotus). Traditional discriminative models trained on real-world data with sparse ground-truth depth can accurately predict metric depth but often produce over-smoothed or low-detail depth maps. Generative models, in contrast, are trained on synthetic data with dense ground truth, generating depth maps with sharp boundaries yet only providing relative depth with low accuracy. Our approach bridges these limitations by integrating metric accuracy with detailed boundary preservation, resulting in depth predictions that are both metrically precise and visually sharp. Our extensive zero-shot evaluations on standard depth estimation benchmarks confirm SharpDepth effectiveness, showing its ability to achieve both high depth accuracy and detailed representation, making it well-suited for applications requiring high-quality depth perception across diverse, real-world environments.
format Preprint
id arxiv_https___arxiv_org_abs_2411_18229
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation
Pham, Duc-Hai
Do, Tung
Nguyen, Phong
Hua, Binh-Son
Nguyen, Khoi
Nguyen, Rang
Computer Vision and Pattern Recognition
We propose SharpDepth, a novel approach to monocular metric depth estimation that combines the metric accuracy of discriminative depth estimation methods (e.g., Metric3D, UniDepth) with the fine-grained boundary sharpness typically achieved by generative methods (e.g., Marigold, Lotus). Traditional discriminative models trained on real-world data with sparse ground-truth depth can accurately predict metric depth but often produce over-smoothed or low-detail depth maps. Generative models, in contrast, are trained on synthetic data with dense ground truth, generating depth maps with sharp boundaries yet only providing relative depth with low accuracy. Our approach bridges these limitations by integrating metric accuracy with detailed boundary preservation, resulting in depth predictions that are both metrically precise and visually sharp. Our extensive zero-shot evaluations on standard depth estimation benchmarks confirm SharpDepth effectiveness, showing its ability to achieve both high depth accuracy and detailed representation, making it well-suited for applications requiring high-quality depth perception across diverse, real-world environments.
title SharpDepth: Sharpening Metric Depth Predictions Using Diffusion Distillation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2411.18229