Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Robust Error-Resistant View Selection Method for 3D Reconstruction

Shaojie Zhang, Yinghui Wang, Bin Nan, Wei Li, Jinlong Yang, Tao Yan, Yukai Wang, Liangyi Huang, Mingfeng Wang, Ibragim R. Atadjanov

TL;DR

This work tackles the problem that small-baseline view selections in Structure-from-Motion introduce triangulation uncertainty. It introduces an error-resistant view selection framework that quantifies triangulation resistance for each camera baseline and builds an error resistance matrix to guide candidate-view sets, followed by a recursive completion to ensure full participation of all views. The method is integrated with COLMAP and evaluated on TUM and DTU datasets, achieving significant reductions in average reprojection error (~28-30%) and modest gains in absolute trajectory error (~4-5%). The approach improves reconstruction robustness and efficiency, offering a practical enhancement for multi-view 3D reconstruction pipelines.

Abstract

To address the issue of increased triangulation uncertainty caused by selecting views with small camera baselines in Structure from Motion (SFM) view selection, this paper proposes a robust error-resistant view selection method. The method utilizes a triangulation-based computation to obtain an error-resistant model, which is then used to construct an error-resistant matrix. The sorting results of each row in the error-resistant matrix determine the candidate view set for each view. By traversing the candidate view sets of all views and completing the missing views based on the error-resistant matrix, the integrity of 3D reconstruction is ensured. Experimental comparisons between this method and the exhaustive method with the highest accuracy in the COLMAP program are conducted in terms of average reprojection error and absolute trajectory error in the reconstruction results. The proposed method demonstrates an average reduction of 29.40% in reprojection error accuracy and 5.07% in absolute trajectory error on the TUM dataset and DTU dataset.

A Robust Error-Resistant View Selection Method for 3D Reconstruction

TL;DR

This work tackles the problem that small-baseline view selections in Structure-from-Motion introduce triangulation uncertainty. It introduces an error-resistant view selection framework that quantifies triangulation resistance for each camera baseline and builds an error resistance matrix to guide candidate-view sets, followed by a recursive completion to ensure full participation of all views. The method is integrated with COLMAP and evaluated on TUM and DTU datasets, achieving significant reductions in average reprojection error (~28-30%) and modest gains in absolute trajectory error (~4-5%). The approach improves reconstruction robustness and efficiency, offering a practical enhancement for multi-view 3D reconstruction pipelines.

Abstract

To address the issue of increased triangulation uncertainty caused by selecting views with small camera baselines in Structure from Motion (SFM) view selection, this paper proposes a robust error-resistant view selection method. The method utilizes a triangulation-based computation to obtain an error-resistant model, which is then used to construct an error-resistant matrix. The sorting results of each row in the error-resistant matrix determine the candidate view set for each view. By traversing the candidate view sets of all views and completing the missing views based on the error-resistant matrix, the integrity of 3D reconstruction is ensured. Experimental comparisons between this method and the exhaustive method with the highest accuracy in the COLMAP program are conducted in terms of average reprojection error and absolute trajectory error in the reconstruction results. The proposed method demonstrates an average reduction of 29.40% in reprojection error accuracy and 5.07% in absolute trajectory error on the TUM dataset and DTU dataset.
Paper Structure (18 sections, 4 equations, 6 figures, 3 tables)

This paper contains 18 sections, 4 equations, 6 figures, 3 tables.

Table of Contents

  1. Introduction
  2. RELATED WORK
  3. METHODOLOGY
  4. Triangulation Error Resistance Calculation
  5. Camera Baseline Error Resistance Mechanism
  6. Calculation of Error Resistance Values
  7. Construction of Error Resistance Matrix
  8. View selection
  9. Determination of Next View Set
  10. View Check and Completion
  11. Experimental and Result Analysis
  12. Evaluation Metrics
  13. Average Reprojection Error
  14. Absolute Trajectory Error
  15. Dataset
  16. ...and 3 more sections

Figures (6)

  • Figure 1: Methodological Framework.
  • Figure 2: Relationship between baseline and triangulation.
  • Figure 3: Schematic Diagram of Triangulation Error Resistance Principle.
  • Figure 4: Reprojection error results on the TUM dataset.
  • Figure 5: Absolute trajectory error results based on the TUM dataset.
  • ...and 1 more figures