Geometry constrained correlation adjustment for stereo reconstruction in 3D optical deformation measurements

Area of Science:

• Computer Vision
• Optical Measurement

Background:

• Accurate 3D shape recovery is crucial for optical deformation measurements.
• Existing methods often rely on triangulation and stereo correspondences, which can be limiting.

Purpose of the Study:

• To propose a novel method for direct depth recovery from binocular vision systems for 3D deformation estimation.
• To overcome limitations of existing stereo correspondence-based techniques.

Main Methods:

• Directly recover depth information from stereo images.
• Parameterize reprojection error by depth and local intensity dissimilarity.
• Formulate a correlation adjustment model to estimate depth by minimizing error.
• Employ Gauss-Newton linearization for solving the model.

Main Results:

• The proposed method accurately and precisely estimates 3D deformation.
• Experimental validation includes rigid translation and bending deformation measurements.
• Demonstrated effectiveness in recovering geometric shape from images.

Conclusions:

• The developed method offers a robust approach for 3D deformation estimation.
• Direct depth recovery provides an accurate alternative to traditional triangulation methods.
• Further developments are in progress.