Acknowledgement
This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT) (No. 2022R1A2C200686411).
Annual Conference of KIPS (한국정보처리학회:학술대회논문집)
- 2022.11a
- /
- Pages.646-647
- /
- 2022
- /
- 2005-0011(pISSN)
- /
- 2671-7298(eISSN)
DOI QR Code
Transformer-based dense 3D reconstruction from RGB images
RGB 이미지에서 트랜스포머 기반 고밀도 3D 재구성
- Xu, Jiajia (Department of Multimedia Engineering, Dongguk University-Seoul) ;
- Gao, Rui (Department of Multimedia Engineering, Dongguk University-Seoul) ;
- Wen, Mingyun (Department of Multimedia Engineering, Dongguk University-Seoul) ;
- Cho, Kyungeun (Department of Multimedia Engineering, Dongguk University-Seoul)
- Published : 2022.11.21
Abstract
Multiview stereo (MVS) 3D reconstruction of a scene from images is a fundamental computer vision problem that has been thoroughly researched in recent times. Traditionally, MVS approaches create dense correspondences by constructing regularizations and hand-crafted similarity metrics. Although these techniques have achieved excellent results in the best Lambertian conditions, traditional MVS algorithms still contain a lot of artifacts. Therefore, in this study, we suggest using a transformer network to accelerate the MVS reconstruction. The network is based on a transformer model and can extract dense features with 3D consistency and global context, which are necessary to provide accurate matching for MVS.
Keywords