PWL 11/2019: Dense Human Pose Estimation In The Wild

We open this new season of Papers We Love Barcelona with a really hot topic: Computer vision.

We count with an expert in the topic: Carlos Andujar from the Universtat Politecnica de Catalunya. Carlos will discuss the paper DensePose: Dense Human Pose Estimation In The Wild from INRIA and Facebook AI Research. The paper can be download here although there is no need to know anything about the topic before the talk.

Talk: DensePose: Dense Human Pose Estimation In The Wild

Dense human pose estimation is defined in this paper like the correspondences between an RGB image and a surface-based representation of the human body. Carlos will explain and discuss the last tecniques in this area like CNN-Based systems and fully-convunational networks from the point of view of a researcher in Graphics.

Supplementary materials, data, code, and videos from the paper are provided on the project page http://densepose.org

Speaker: Carlos Andújar Gran
Carlos is an associate professor in the Computer Science Dept. of the Universitat Politècnica de Catalunya - BarcelonaTech. He is member of the ViRVIG - Research Center for Visualization, Virtual Reality and Graphics Interaction. He is also member of the Eurographics Association.
His current research interests are: Computer Graphics, 3D modeling and animation, Virtual and augmented reality and 3D user interfaces.