The field of 3D reconstruction and deformation is rapidly advancing, with a focus on developing methods that can accurately and robustly reconstruct 3D objects and scenes from various types of input data. Recent research has explored the use of deep learning techniques, such as neural networks and generative models, to improve the accuracy and efficiency of 3D reconstruction algorithms. Additionally, there is a growing interest in developing methods that can handle complex deformations and dynamic scenes, such as those found in videos or interactive applications. Notable papers in this area include: PAD3R, which presents a method for reconstructing deformable 3D objects from casually captured monocular videos, and MPMAvatar, which introduces a framework for creating 3D human avatars with accurate and robust physics-based dynamics. These papers demonstrate significant advancements in the field, enabling more accurate and robust 3D reconstruction and deformation, and paving the way for applications in areas such as robotics, computer vision, and graphics.