Seamless texture map generation from multiple images

Authors:
Robert Valkenburg;Nawar Alwesh
Affiliations:
Industrial Research Limited, Auckland, New Zealand;Industrial Research Limited, Auckland, New Zealand
Venue:
Proceedings of the 27th Conference on Image and Vision Computing New Zealand
Year:
2012

Citing 3
Cited 0

Multiple color texture map fusion for 3D models

Pattern Recognition Letters
Technical Section: Masked photo blending: Mapping dense photographic data set on high-resolution sampled 3D models

Computers and Graphics
KinectFusion: Real-time dense surface mapping and tracking

ISMAR '11 Proceedings of the 2011 10th IEEE International Symposium on Mixed and Augmented Reality

Quantified Score

Hi-index	0.00

Visualization

Abstract

This paper presents an algorithm to reduce the appearance of seams in textured 3D models where the texture map is generated from images captured from the scene. Unless images are acquired under ideal diffuse lighting conditions the textured model will exhibit seams between faces due to image illumination variations. The algorithm minimises the perceptual variation of the colour at each vertex of the 3D mesh in a global optimisation. This optimisation gives rise to a large sparse linear system. The algorithm is used to generate a small beach scene model and shows a significant improvement in the qualitative appearance of the model.