Refinement of Monocular Depth Maps via Multi-View Differentiable Rendering

dc.contributor.authorFink, Lauraen_US
dc.contributor.authorFranke, Linusen_US
dc.contributor.authorEgger, Bernharden_US
dc.contributor.authorKeinert, Joachimen_US
dc.contributor.authorStamminger, Marcen_US
dc.contributor.editorEgger, Bernharden_US
dc.contributor.editorGünther, Tobiasen_US
dc.date.accessioned2025-09-24T10:37:18Z
dc.date.available2025-09-24T10:37:18Z
dc.date.issued2025
dc.description.abstractAccurate depth estimation is at the core of many applications in computer graphics, vision, and robotics. Current state-ofthe- art monocular depth estimators, trained on extensive datasets, generalize well but lack 3D consistency needed for many applications. In this paper, we combine the strength of those generalizing monocular depth estimation techniques with multiview data by framing this as an analysis-by-synthesis optimization problem to lift and refine such relative depth maps to accurate error-free depth maps. After an initial global scale estimation through structure-from-motion point clouds, we further refine the depth map through optimization enforcing multi-view consistency via photometric and geometric losses with differentiable rendering of the meshed depth map. In a two-stage optimization, scaling is further refined first, and afterwards artifacts and errors in the depth map are corrected via nearby-view photometric supervision. Our evaluation shows that our method is able to generate detailed, high-quality, view consistent, accurate depth maps, also in challenging indoor scenarios, and outperforms state-of-the-art multi-view depth reconstruction approaches on such datasets. Project page and source code can be found at https://lorafib.github.io/ref_depth/.en_US
dc.description.sectionheadersNeural and Differentiable Rendering
dc.description.seriesinformationVision, Modeling, and Visualization
dc.identifier.doi10.2312/vmv.20251232
dc.identifier.isbn978-3-03868-294-3
dc.identifier.pages10 pages
dc.identifier.urihttps://doi.org/10.2312/vmv.20251232
dc.identifier.urihttps://diglib.eg.org/handle/10.2312/vmv20251232
dc.publisherThe Eurographics Associationen_US
dc.rightsAttribution 4.0 International License
dc.rights.urihttps://creativecommons.org/licenses/by/4.0/
dc.subjectCCS Concepts: Computing methodologies → Computer vision problems; Rasterization
dc.subjectComputing methodologies → Computer vision problems
dc.subjectRasterization
dc.titleRefinement of Monocular Depth Maps via Multi-View Differentiable Renderingen_US
Files
Original bundle
Now showing 1 - 3 of 3
Loading...
Thumbnail Image
Name:
vmv20251232.pdf
Size:
9.66 MB
Format:
Adobe Portable Document Format
No Thumbnail Available
Name:
paper1037_2.mp4
Size:
59.37 MB
Format:
Video MP4
Loading...
Thumbnail Image
Name:
paper1037_3.pdf
Size:
40.21 MB
Format:
Adobe Portable Document Format
Collections