Unsupervised Learning of Shape and Pose with Differentiable Point Clouds

"AI4Animation
This project explores the opportunities of deep learning for character animation and control as part of my Ph.D. research at the University of Edinburgh in the School of Informatics, supervised by Taku Komura..."

"...Say goodbye to flatland. We show how curriculum+RL can learn policies to traverse difficult step sequences and variable terrain (rewards only; no example data)..."

YOLO5 is here! https://blog.roboflow.ai/yolov5-is-here/

DisentangledFaceGAN

ApartmentPlan 0.2 StyleGAN2

ApartmentPlan 0.3 StyleGAN2

"Q. Can we find winning lottery tickets, or sparse trainable deep networks at initialization without ever looking at data?

A. Yes, by conserving "Synaptic Flow" via our new SynFlow algorithm..."

Dance Revolution: Long Sequence Dance Generation
with Music via Curriculum Learning

"The code and data for our previous animation research on character-scene interactions is online. Mocap available as annotated Unity assets. Includes locomotion, sitting, carrying, opening doors, avoiding obstacles and more..."

XiaoiceSing: A High-Quality and Integrated Singing Voice Synthesis System

Thinking Machines
"Thinking Machines радикально преобразует рабочий процесс дизайнеров, пересматривая концепцию «персональных компьютеров» в эпоху нейронных сетей."

Thinking Machines - это семестровый проект в Media Design Master, HEAD - Genève, в котором студенты будут пересматривать концепцию «персональных компьютеров» в эпоху нейронных сетей и исследовать, как создание персонализированных помощников может изменить их творческий рабочий процесс.

The code of the research "PIFuHD", which was adopted by Oral and which realizes high resolution full 3D of person from one image, has been released on github

Coherent Reconstruction of Multiple Humans from a Single Image