MEDeA: Multi-view Efficient Depth Adjustment

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Artemyev, Mikhail, Vorontsova, Anna, Sokolova, Anna, Limonov, Alexander
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910491758034944
author Artemyev, Mikhail
Vorontsova, Anna
Sokolova, Anna
Limonov, Alexander
author_facet Artemyev, Mikhail
Vorontsova, Anna
Sokolova, Anna
Limonov, Alexander
contents The majority of modern single-view depth estimation methods predict relative depth and thus cannot be directly applied in many real-world scenarios, despite impressive performance in the benchmarks. Moreover, single-view approaches cannot guarantee consistency across a sequence of frames. Consistency is typically addressed with test-time optimization of discrepancy across views; however, it takes hours to process a single scene. In this paper, we present MEDeA, an efficient multi-view test-time depth adjustment method, that is an order of magnitude faster than existing test-time approaches. Given RGB frames with camera parameters, MEDeA predicts initial depth maps, adjusts them by optimizing local scaling coefficients, and outputs temporally-consistent depth maps. Contrary to test-time methods requiring normals, optical flow, or semantics estimation, MEDeA produces high-quality predictions with a depth estimation network solely. Our method sets a new state-of-the-art on TUM RGB-D, 7Scenes, and ScanNet benchmarks and successfully handles smartphone-captured data from ARKitScenes dataset.
format Preprint
id arxiv_https___arxiv_org_abs_2406_12048
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MEDeA: Multi-view Efficient Depth Adjustment
Artemyev, Mikhail
Vorontsova, Anna
Sokolova, Anna
Limonov, Alexander
Computer Vision and Pattern Recognition
Artificial Intelligence
The majority of modern single-view depth estimation methods predict relative depth and thus cannot be directly applied in many real-world scenarios, despite impressive performance in the benchmarks. Moreover, single-view approaches cannot guarantee consistency across a sequence of frames. Consistency is typically addressed with test-time optimization of discrepancy across views; however, it takes hours to process a single scene. In this paper, we present MEDeA, an efficient multi-view test-time depth adjustment method, that is an order of magnitude faster than existing test-time approaches. Given RGB frames with camera parameters, MEDeA predicts initial depth maps, adjusts them by optimizing local scaling coefficients, and outputs temporally-consistent depth maps. Contrary to test-time methods requiring normals, optical flow, or semantics estimation, MEDeA produces high-quality predictions with a depth estimation network solely. Our method sets a new state-of-the-art on TUM RGB-D, 7Scenes, and ScanNet benchmarks and successfully handles smartphone-captured data from ARKitScenes dataset.
title MEDeA: Multi-view Efficient Depth Adjustment
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2406.12048