Network Morphism
Mining 3D Key-Pose-Motifs for Action Recognition
Deep Neural Networks
We study how to morph a well-trained neural network to a new one, and how to design advanced deep neural networks.
Faculty Summit 2016 – Spatial Audio for Augmented and Virtual Reality
Head mounted displays for virtual and augmented reality are a hot topic for research and product development. Integral part of these devices is the spatial audio rendering system. Unlike the vision, where humans have approximately…
Microsoft Research Faculty Summit opens
By Harold Javid, General Chair, Faculty Summit Microsoft Research’s annual Faculty Summit opened Wednesday with a series of talks about how technological innovation can benefit both business and society. In a fireside chat with Jeannette M. Wing,…
Safe Autonomous Flight Everywhere
The goal of this project is to build a generalized vehicle-agnostic system that can provide safety guarantees in the face of an uncertain world. We are defining what “safe” means, and building a theoretical framework…
Efficient and Precise Interactive Hand Tracking through Joint, Continuous Optimization of Pose and Correspondences
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…