MrNeRF (@janusch_patas)
2025-07-22 | โค๏ธ 128 | ๐ 19
Advances in Feed-Forward 3D Reconstruction and View Synthesis: A Survey
Abstract (excerpt): This survey offers a comprehensive review of feed-forward techniques for u3D reconstruction and view synthesis. It provides a taxonomy based on underlying representation architectures, including point cloud, 3D Gaussian Splatting (3DGS), Neural Radiance Fields (NeRF), etc.
We examine key tasks such as pose-free reconstruction, dynamic 3D reconstruction, and 3D-aware image and video synthesis, highlighting their applications in digital humans, SLAM, robotics, and beyond. Additionally, we review commonly used datasets with detailed statistics and evaluation protocols for various downstream tasks.
We conclude by discussing open research challenges and promising directions for future work, emphasizing the potential of feed-forward approaches to advance the state of the art in 3D vision.
๐ Related
See similar notes in domain-vision-3d, domain-robotics