Skip to content

Read-only version: For security reasons, the Research Information System (FIS) has been temporarily replaced by this read-only version. All public content and full texts remain available; login, editing, submission and statistics are not available. Questions: fis@uni-bamberg.de

Home › Publications › Multi-stage mask-aware Depth Enhancement for RGB–IR–stereo Fusion on historic Timber Surfaces

Publication

Multi-stage mask-aware Depth Enhancement for RGB–IR–stereo Fusion on historic Timber Surfaces

Pan, Junquan; Chizhova, Maria; Hess, Mona; u. a. (2026): Multi-stage mask-aware Depth Enhancement for RGB–IR–stereo Fusion on historic Timber Surfaces, in: ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Katlenburg-Lindau: Copernicus Publications, S. 217–226, doi: 10.5194/isprs-annals-xi-1-2026-217-2026.

Details

Faculty/Chair
Faculty/Professorship
Author
Luhmann, Thomas
Maiwald, Ferdinand
Title of the Journal
ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
ISSN
2194-9050
Conference
XXV ISPRS Congress 2026 “From Imagery to Understanding”, Commission I, 4–11 July 2026, Toronto, Canada
Publisher Information
Copernicus Publications : Katlenburg-Lindau
Year of publication
2026
Volume
XI-1-2026
Language
English
DOI
URL
Abstract

This paper presents a mask-aware multi-stage depth enhancement framework for digital documentation of historical timber surfaces using RGB–Stereo-IR fusion. Accurate geometric recording of aged wood features such as wooden knots remains challenging due to uneven illumination and weak texture. The proposed pipeline, which aims to stabilise depth geometry under uneven illumination and low-texture surface conditions, integrates object detection, instance segmentation and confidence-guided depth refinement across three stages: (A) TV(total variation)-regularized mask-aware refinement, (B) confidence-weighted multi-view fusion, and (C) patch-based stereo reconstruction. Experiments on historical timber beams under varying illumination demonstrate improved depth completeness and geometric consistency, achieving a residual standard deviation below 0.6 mm in bright scenes and stable reconstruction in low-light conditions. The framework offers a practical solution for depth reconstruction of cultural heritage timber, supporting more reliable feature detection and analysis.

GND Keywords
Bauholz
Oberfläche
Denkmalkunde
Digitalisierung
Dokumentation
Keywords
Object detection
Instance segmentation
Mask-aware depth enhancement
RGB-IR–Stereo fusion
Heritage timber
Damage assessment
DDC Classification
920 Biography, genealogy, heraldic
RVK Classification
LH 60360
Peer Reviewed
yes
International Distribution
yes
Open Access Journal
yes
Type
Conferenceobject
URI
Activation date
2026-07-17
Project(s)

Permalink: https://fis.uni-bamberg.de/handle/uniba/116223 · Data as JSON · Cite: BibTeX · RIS

Data as of: 2026-11-16

Restricted read-only access: login, editing and submission are temporarily disabled.