High Quality Reconstruction
High-quality reconstruction aims to generate detailed and accurate representations of objects or scenes from incomplete or noisy data, a crucial task across diverse fields. Current research focuses on improving reconstruction quality and speed using neural networks, particularly neural radiance fields (NeRFs) and Gaussian splatting, often incorporating techniques like 3D geometry guidance, motion compensation, and adaptive sampling strategies to address challenges such as artifacts, sparse data, and computational cost. These advancements have significant implications for various applications, including medical imaging, autonomous driving, and 3D modeling, enabling more accurate and efficient analysis and visualization of complex data.
44papers
Papers
April 8, 2025
April 7, 2025
February 12, 2025
January 14, 2025
December 17, 2024
December 10, 2024
November 1, 2024
October 27, 2024
October 21, 2024
August 29, 2024