Title
Real time 360° video stitching and streaming.
Abstract
In this paper we propose a real time 360° video stitching and streaming processing methodology focused on GPU. The solution creates a scalable solution for large resolutions, such as 4K and 8K per camera, and supports broadcasting solutions with cloud architectures. The methodology uses a group of deformable meshes, processed using OpenGL (GLSL) and the final image combine the inputs using a robust pixel shader. Moreover, the result can be streamed to a cloud service using h.264 encoding with nVEnc GPU encoding. Finally, we present some results.
Year
DOI
Venue
2016
10.1145/2945078.2945148
SIGGRAPH Posters
Field
DocType
Citations 
Computer vision,Broadcasting,Polygon mesh,Virtual reality,Computer graphics (images),Computer science,Pixel,Artificial intelligence,Shader,OpenGL,Encoding (memory),Cloud computing
Conference
4
PageRank 
References 
Authors
0.45
3
5
Name
Order
Citations
PageRank
Rodrigo M. A. Silva140.45
Bruno Feijó220029.85
Pablo B. Gomes340.45
Thiago Frensh440.45
Daniel Monteiro540.79