SOTAVerified

Novel View Synthesis

Synthesize a target image with an arbitrary target camera pose from given source images and their camera poses.

See Wiki for more introductions.

The Synthesis method include: NeRF, MPI and so on.

( Image credit: Multi-view to Novel view: Synthesizing novel views with Self-Learned Confidence )

Papers

Showing 1–10 of 1441 papers

TitleStatusHype
Physically Based Neural LiDAR ResimulationCode0
Cameras as Relative Positional Encoding—0
MoVieS: Motion-Aware 4D Dynamic View Synthesis in One Second—0
LighthouseGS: Indoor Structure-aware 3D Gaussian Splatting for Panorama-Style Mobile Captures—0
Reflections Unlock: Geometry-Aware Reflection Disentanglement in 3D Gaussian Splatting for Photorealistic Scenes Rendering—0
Outdoor Monocular SLAM with Global Scale-Consistent 3D Gaussian Pointmaps—0
Refine Any Object in Any SceneCode1
VoteSplat: Hough Voting Gaussian Splatting for 3D Scene Understanding—0
UnMix-NeRF: Spectral Unmixing Meets Neural Radiance FieldsCode1
DBMovi-GS: Dynamic View Synthesis from Blurry Monocular Video via Sparse-Controlled Gaussian Splatting—0
Show:102550
← PrevPage 1 of 145Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TensoRF + NeRFLiXPSNR27.39—Unverified
2JAXNeRFPSNR26.92—Unverified
3K-Planes (hybrid)PSNR26.92—Unverified
4Plenoxels + NeRFLiXPSNR26.9—Unverified
5K-Planes (explicit)PSNR26.78—Unverified
6TensoRFPSNR26.73—Unverified
7NeRFPSNR26.5—Unverified
8PlenoxelsPSNR26.29—Unverified
9HyperReelPSNR26.2—Unverified
10MobileNeRFPSNR25.91—Unverified