Generative Network
Generative networks are artificial neural networks designed to learn complex data distributions and generate new samples resembling the training data. Current research focuses on improving the quality and diversity of generated outputs, addressing issues like mode collapse and achieving better control over generation through techniques such as conditional generation and network bending. These advancements are driving progress in diverse fields, including image synthesis, 3D modeling, and data augmentation for tasks like anomaly detection and few-shot learning, ultimately enhancing the capabilities of various machine learning applications.
Papers
Focus on Content not Noise: Improving Image Generation for Nuclei Segmentation by Suppressing Steganography in CycleGAN
Jonas Utz, Tobias Weise, Maja Schlereth, Fabian Wagner, Mareike Thies, Mingxuan Gu, Stefan Uderhardt, Katharina Breininger
BEVControl: Accurately Controlling Street-view Elements with Multi-perspective Consistency via BEV Sketch Layout
Kairui Yang, Enhui Ma, Jibin Peng, Qing Guo, Di Lin, Kaicheng Yu