Volume Rendering
Volume rendering is a computer graphics technique for visualizing 3D data by projecting the volume's internal structure onto a 2D image. Current research focuses on improving rendering speed and quality, particularly for real-time applications, through advancements in model architectures like Gaussian splatting and neural radiance fields (NeRFs). These methods are being refined to handle complex scenes, dynamic objects, and challenging conditions like sparse data or scattering media, leading to more efficient and accurate representations for various applications. The resulting improvements have significant implications for fields like medical imaging, virtual reality, and autonomous driving, enabling more realistic and interactive visualizations.
Papers
VolRecon: Volume Rendering of Signed Ray Distance Functions for Generalizable Multi-View Reconstruction
Yufan Ren, Fangjinhua Wang, Tong Zhang, Marc Pollefeys, Sabine Süsstrunk
Real-Time Neural Light Field on Mobile Devices
Junli Cao, Huan Wang, Pavlo Chemerys, Vladislav Shakhrai, Ju Hu, Yun Fu, Denys Makoviichuk, Sergey Tulyakov, Jian Ren
SteerNeRF: Accelerating NeRF Rendering via Smooth Viewpoint Trajectory
Sicheng Li, Hao Li, Yue Wang, Yiyi Liao, Lu Yu