Research
Fully articulated hand tracking promises to enable fundamentally new interactions with virtual and augmented worlds, but the limited accuracy and efficiency of current systems has prevented widespread adoption. Today’s dominant paradigm uses machine learning for…
We contribute a new pipeline for live multi-view performance capture, generating temporally coherent high-quality reconstructions in real-time. Our algorithm supports both incremental reconstruction, improving the surface estimation over time, as well as parameterizing the nonrigid…
MSR Image Recognition Challenge (IRC) @ IEEE ICME 2016 (past) ICME 2016 Image Recognition Grand Challenge Session: Time: 10:00-11:40, Wednesday, July 13, 2016 Room: Grand III Deep Multi-Context Network for Fine-Grained Visual Recognition Xinyu Ou1,2,3,…
By Guobin Wu, Senior Research Program Manager, Microsoft Research Asia What does it take to make it to the “decisive round for the finest few” in the Beauty of Programming Contest 2016? Earlier this year,…
Spatial sound reproduction methods for recorded sound scenes are an active field of research, in parallel with the evolving vision-related or multi-modal technologies that aim to deliver a new generation of immersive multimedia content to…
We have developed an efficient entropy coder for integer value data. We refer to this compression algorithm as a Run-Length Golomb-Rice (RLGR) coder. The RLGR coder is very simple to implement, and uses backward adaptation…
The still image compression format now known as JPEG XR has its roots at Microsoft Research. It started as the Progressive Transform Codec (PTC), which we designed back in 1999 as an alternative to JPEG 2000.…
