Abstract
In this paper, we introduce a hybrid and user-friendly 3D reconstruction system for objects, which seamlessly integrates a deep neural network model for 3D shape reconstruction and a multi-band image blending algorithm for realistic texturing into a single encapsulated application, dubbed as 3D-Producer. Compared to expensive laser scanning devices, our 3D-Producer offers an economical and practical solution, where a textured 3D model for a target object can be easily created by general users with mobile devices, a physical turntable, and our application. An illustration video can be found here (https://github.com/winnechan/3D-Producer/blob/main/3D-Producer-Demo-480P.mp4 ).
Author supplied keywords
Cite
CITATION STYLE
Chen, J., Zhang, Y., Zhang, Z., Pan, Y., & Yao, T. (2022). 3D-Producer: A Hybrid and User-Friendly 3D Reconstruction System. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 13606 LNAI, pp. 526–531). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-031-20503-3_43
Register to see more suggestions
Mendeley helps you to discover research relevant for your work.
