Dynamic Facial Analysis: From Bayesian Filtering to Recurrent Neural Network

Facial analysis in videos, including head pose estimation and facial landmark localization, is key for many applications such as facial animation capture, human activity recognition, and human-computer interaction. In this paper, we propose to use a recurrent neural network (RNN) for joint estimation and tracking of facial features in videos. We are inspired by the fact that the computation performed in an RNN bears resemblance to Bayesian filters, which have been used for tracking in many previous methods for facial analysis from videos.

Jacob Munkberg

Jacob Munkberg is a principal research scientist in NVIDIA's real-time rendering research group.

Prior to joining NVIDIA Research in 2016, Jacob worked as senior research scientist in Intel’s Advanced Rendering Technology team. He joined Intel in 2008 via Intel's acquisition of the computer graphics startup company, Swiftfoot Graphics, specializing in culling technology and efficient multi-view graphics. Jacob received his Ph.D. in computer science from Lund University and his M.S. in engineering physics from Chalmers University of Technology.

Tomas Akenine-Möller

Tomas Akenine-Möller joined NVIDIA Research in 2016, previously working at Intel and as a professor in computer graphics at Lund University, where he founded and built the computer graphics group. His expertise is in real-time rendering, ray tracing, and graphics hardware. In 2003, he co-authored the first Swedish SIGGRAPH papers ever, one on mobile graphics and another on soft shadows.

Petrik Clarberg

Petrik Clarberg joined the Real-Time Rendering team at NVIDIA Research in 2016. His current work is focused on systems and algorithmic research for real-time path tracing and new appearance models. Petrik's passion for graphics started with the 1990s demo scene. Before joining NVIDIA, he worked with graphics research in the industry for many years, co-founded a startup, and completed PhD studies at Lund University, Sweden.

NVIDIA Papers Win First and Second Best Paper Awards at HPG 2016

Date

Best paper:

Anton S. Kaplanyan, Stephen Hill, Anjul Patney, and Aaron Lefohn, "Filtering Distributions of Normals for Shading Antialiasing"

Second Best Paper:

Chris Wyman, "Exploring and Expanding the Continuum of OIT Algorithms"

2017 Grad Fellows

NVIDIA Graduate Fellowship Results for 2017

We are excited to announce the 2017 NVIDIA Graduate Fellowship recipients!

We know that there is incredibly important work taking place at universities worldwide, and the NVIDIA Graduate Fellowship Program allows us to demonstrate our commitment to academia in supporting research that spans all areas of computing innovation. In particular this year, emphasis was given to students pushing the envelope in artificial intelligence, deep neural networks, autonomous vehicles, and related fields.