Deformable Nature
Deformable nature research focuses on modeling and manipulating objects that change shape, a significant challenge in computer vision and robotics. Current efforts concentrate on developing efficient algorithms, such as deformable transformers and Gaussian splatting, to represent and track these deformations in various contexts, including video object segmentation, 3D reconstruction, and robotic manipulation of textiles and other flexible materials. This research is crucial for advancing applications ranging from automated agricultural tasks and security screening to medical image analysis and high-fidelity virtual reality. Improved modeling of deformable objects promises significant advancements across numerous fields.
Papers
November 6, 2024
October 30, 2024
July 1, 2024
June 5, 2024
May 11, 2024
April 4, 2024
November 24, 2023
November 13, 2023
September 22, 2023
June 5, 2023
March 24, 2023
November 7, 2022
September 22, 2022
July 22, 2022
April 14, 2022