Deformable Shape Reconstruction

Overview

Deformable shape reconstruction studies how to recover the changing geometry of objects such as tissue, cloth, hands, and soft robots from incomplete observations. Methods use deformable meshes and templates, neural implicit fields and Gaussian splats, or sensor-based shape estimates, often combining visual evidence with physical or geometric constraints. The work supports applications including surgical navigation, robotic manipulation, animation, and simulation, with growing attention to real-time reconstruction and prediction. A central challenge is recovering accurate, physically plausible shapes despite occlusion, sparse views, camera motion, and differences between training objects and objects encountered in practice.

Latest papers

CardsList

62 papers7 papers in the last four weeks, up 40% on the four weeks before. 0.1% of all new papers.

  1. EAvatar: Expression-Aware Head Avatar Reconstruction with Generative Geometry Priors

    Aug 19, 2025Shikun Zhang, Cunjian Chen, Yiqun Wang +13D Face Reconstruction3D Gaussian Splatting

  2. Online camera-pose-free stereo endoscopic tissue deformation recovery with tissue-invariant vision-biomechanics consistency

    Jun 24, 2025Jiahe Chen, Naoki Tomii, Ichiro Sakuma +1Endoscopy3D Reconstruction