Abstract: While computer vision and deep learning research has made tremendous progress, most success stories are limited to specific domains in which lots of carefully-labeled data can be unambiguously and easily acquired. In this talk, I will present my research on creating systems that can learn to understand visual data with minimal human supervision. In particular, I’ll focus on algorithms that can discover meaningful disentangled representations for controllable image generation, and learn 3D aware visual representations in a self-supervised manner.
Orchard View Room, Virtual
Yong Jae Lee