There are wide varieties of applications of 3D modeling from real images, which can generate virtual view images by rendering the input images onto the 3D shape model. Because the virtual viewpoint is not generally at the same point, the synthesized image at virtual view does not have sufficient resolution. For improving this resolution problem, we propose a method for improving the rendered texture resolution. For each vertex of the 3D shape model, the position of the vertex is adjusted for improving the texture of the neighboring triangle patches. For the improvement, the position of vertex is adjusted by comparing the observation (input image) with projected image of model texture. Repeating the visit of every vertex, not only super resolved texture but also more accurate 3D patch model can be obtained. By performing the experiments using synthesized multiple view images of 51 calibrated cameras around the object scene and real VTR image sequence, it is demonstrated that the proposed method can successfully provide super resolved texture image with improved shape 3D model.
|ジャーナル||Proceedings of the IEEE International Conference on Systems, Man and Cybernetics|
|出版物ステータス||Published - 2000 12 1|
|イベント||2000 IEEE Interantional Conference on Systems, Man and Cybernetics - Nashville, TN, USA|
継続期間: 2000 10 8 → 2000 10 11
ASJC Scopus subject areas
- Control and Systems Engineering
- Hardware and Architecture