… :Neural rendering with implicit neural networks has recently emerged as an attractive proposition
for scene … present Neural Adaptive Scene Tracing (NAScenT), the first neural rendering …

… neural representation method that represents a scene as an octree with a coordinate-based
neural … method that jointly solves multiple neural networks representation and computational …

… the cross-validated results for each scene: For every test scene, we train on all the other
scenes, and output the average PSNR over the whole test scene. We use PSNR as metric …

… propose an adaptive multiscale neural scene representation that fits large-scale 2D and
complex 3D scenes significantly faster and better than existing neural scene representations. …

… a neural ray marching algorithm with learned, adaptive step size to find ray intersections
with scene … Here, we take inspiration from the classic sphere tracing algorithm [67]. Sphere …

… and [47] is that their raytracing is performed on screen space while ours is in a 3D volumetric
space. For a ray r : x = o + tvs, we can trace and intersect it with the neural SDF field. The …

… neural radiance caching method for path-traced global illumination. Our system is designed
to handle fully dynamic scenes… pretraining neural networks to handle novel, dynamic scenes …

… depth maps instead of the full scene models in 3D space. As a … to represent 3D scenes as
implicit neural representations [32… scene and render it into images based on the sphere tracing …

… two scenes with different distributed effects. First, we examine the path traced Sibenik scene
with … The Anim-Blue Sphere scene is a challenging, path-traced scene containing three balls …

… volume path tracing with multiple scattering, temporally stable neural … Additionally, we
integrate neural adaptive sampling to … Our scenes are currently lit from both an HDR environment …