Deformation Field
Deformation fields represent the mapping of points from one space to another, crucial for tasks like image registration and 3D shape modeling. Current research focuses on improving the accuracy, efficiency, and robustness of deformation field estimation, employing diverse approaches such as diffusion models, neural networks (including U-Nets and Transformers), and physics-informed methods. These advancements are driving progress in medical imaging (e.g., more accurate brain and cardiac analysis), computer graphics (e.g., realistic avatar creation and animation), and other fields requiring precise modeling of non-rigid transformations. The ability to accurately and efficiently estimate deformation fields is increasingly important for various applications, improving the precision and interpretability of analyses across multiple disciplines.
Papers
MemWarp: Discontinuity-Preserving Cardiac Registration with Memorized Anatomical Filters
Hang Zhang, Xiang Chen, Renjiu Hu, Dongdong Liu, Gaolei Li, Rongguang Wang
Deformation-Recovery Diffusion Model (DRDM): Instance Deformation for Image Manipulation and Synthesis
Jian-Qing Zheng, Yuanhan Mo, Yang Sun, Jiahua Li, Fuping Wu, Ziyang Wang, Tonia Vincent, Bartłomiej W. Papież