Computer Vision News - May 2021

22 MICCAI Challenge MICCAI for focused discussion on generative modeling in both MIC and CAI. “Deep generative models such as Generative Adversarial Network (GAN) and Variational Auto-Encoder (VAE) combine advanced deep neural networks with classical density estimation (either explicit or implicit) for achieving state-of-the-art results, ” Sandy explains. “GANs are applied to many MIC problems, such as segmentation, registration, synthesis, reconstruction, and denoising, and are the most popular among the deep generative models, but explicit density modeling by VAE has seen a surge of latent space disentanglement in medical imaging literature.” The workshop will host two keynote speakers, Stefanie Speidel and Andreas Maier , who are both great friends of Computer Vision News . Depending on the course of the pandemic in the coming months, MICCAI will either be a hybrid event, allowing both online and in-person participation in Strasbourg, or a virtual meeting. “We would like to do it as a hybrid event if possible, ” Sandy says hopefully.