Guest
Speaker: Ming-Yu Liu, In this talk, I review several recent semantic image synthesis work in the past few years. I put them in the same perspective and illustrate the architecture change. The two main objective functions for training these two models are also discussed and compared. 1. High-resolution image synthesis and semantic manipulation with conditional gans, CVPR'2018 () 2. Semantic Image Synthesis with Spatially-Adaptive Normalization, CVPR'2019 () 3. Taming transformers for high-resolution image synthesis, CVPR'2021 () Tutorial website:
