Skip to content

guikunchen/3DGS-Benchmarks

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 

Repository files navigation

Performance Comparison Tables


Table 1: Comparison of localization methods on Replica

Comparison of localization methods on Replica (static scenes), in terms of absolute trajectory error (ATE, cm).

Method GS Room0 Room1 Room2 Office0 Office1 Office2 Office3 Office4 Average
Vox-Fusion 1.37 4.70 1.47 8.48 2.04 2.58 1.11 2.94 3.09
NICE-SLAM 0.97 1.31 1.07 0.88 1.00 1.06 1.10 1.13 1.06
ESLAM 0.71 0.70 0.52 0.57 0.55 0.58 0.72 0.63 0.63
Point-SLAM 0.61 0.41 0.37 0.38 0.48 0.54 0.69 0.72 0.52
Co-SLAM 0.70 0.95 1.35 0.59 0.55 2.03 1.56 0.72 1.00
Gaussian-SLAM 3.35 8.74 3.13 1.11 0.81 0.78 1.08 7.21 3.27
GSSLAM 0.47 0.43 0.31 0.70 0.57 0.31 0.31 0.31 0.79
GS-SLAM 0.48 0.53 0.33 0.52 0.41 0.59 0.46 0.70 0.50
SplaTAM 0.31 0.40 0.29 0.47 0.27 0.29 0.32 0.55 0.36

Table 2: Comparison of mapping methods on Replica

Comparison of mapping methods on Replica (static scenes), in terms of PSNR, SSIM, and LPIPS.

Method GS Metric Room0 Room1 Room2 Office0 Office1 Office2 Office3 Office4 Average FPS
NICE-SLAM PSNR↑ 22.12 22.47 24.52 29.07 30.34 19.66 22.23 24.94 24.42 0.81
SSIM↑ 0.69 0.76 0.81 0.87 0.89 0.80 0.80 0.86 0.81
LPIPS↓ 0.33 0.27 0.21 0.23 0.18 0.23 0.21 0.20 0.23
Vox-Fusion PSNR↑ 22.39 22.36 23.92 27.79 29.83 20.33 23.47 25.21 24.41 2.17
SSIM↑ 0.68 0.75 0.80 0.86 0.88 0.79 0.80 0.85 0.80
LPIPS↓ 0.30 0.27 0.23 0.24 0.18 0.24 0.21 0.20 0.24
Point-SLAM PSNR↑ 32.40 34.08 35.50 38.26 39.16 33.99 33.48 33.49 35.17 1.33
SSIM↑ 0.97 0.98 0.98 0.98 0.98 0.96 0.96 0.98 0.97
LPIPS↓ 0.11 0.12 0.11 0.10 0.12 0.16 0.13 0.14 0.12
SplaTAM PSNR↑ 32.86 33.89 35.25 38.26 39.17 31.97 29.70 31.81 34.11 -
SSIM↑ 0.98 0.98 0.98 0.98 0.98 0.95 0.95 0.97
LPIPS↓ 0.07 0.10 0.08 0.09 0.09 0.10 0.12 0.15 0.10
GSSLAM PSNR↑ 31.56 32.86 32.59 38.70 41.17 32.36 32.03 32.92 34.27 -
SSIM↑ 0.97 0.97 0.97 0.99 0.99 0.98 0.98 0.97 0.97
LPIPS↓ 0.07 0.07 0.07 0.05 0.03 0.09 0.11 0.11 0.08
GSSLAM PSNR↑ 34.83 36.43 37.49 39.95 42.09 36.24 36.70 36.07 37.50 769
SSIM↑ 0.98 0.98 0.96 0.97 0.98 0.98 0.98 0.96 0.98
LPIPS↓ 0.07 0.08 0.07 0.07 0.06 0.08 0.07 0.10 0.07
Gaussian-SLAM PSNR↑ 34.31 37.28 38.18 43.97 43.56 37.39 36.48 40.19 38.90 -
SSIM↑ 0.99 0.99 0.99 1.00 0.99 0.99 0.99 1.00 0.99 -
LPIPS 0.08 0.07 0.07 0.04 0.04 0.07 0.07 0.07 0.07 -

Table 3: Comparison of reconstruction methods on D-NeRF

Comparison of reconstruction methods on D-NeRF (dynamic scenes), in terms of PSNR, SSIM, and LPIPS.

Method GS PSNR↑ SSIM↑ LPIPS↓
D-NeRF 30.50 0.95 0.07
TiNeuVox-B 32.67 0.97 0.04
KPlanes 31.61 0.97 -
HexPlane-Slim 32.68 0.97 0.02
MSTH 31.34 0.98 0.02
3D GS 23.19 0.93 0.08
4DGS 34.09 0.98 -
4D-GS 34.05 0.98 0.02
GaGS 37.36 0.99 0.01
D-3DGS 39.51 0.99 0.02

Table 4: Comparison of reconstruction methods on ZJU-MoCap

Comparison of reconstruction methods on ZJU-MoCap (avatar), in terms of PSNR, SSIM, and LPIPS*. The numbers of non-GS methods are taken from GART.

Method GS PSNR↑ SSIM↑ LPIPS↓*
NeuralBody 29.03 0.96 42.47
AnimNeRF 29.77 0.96 46.89
PixelNeRF 24.71 0.89 121.86
NHP 28.25 0.95 64.77
HumanNeRF 30.66 0.97 33.38
Instant-NVR 31.01 0.97 38.45
GauHuman 31.34 0.97 30.51
3DGS-Avatar 30.61 0.97 29.58
GART 32.22 0.98 29.21

Table 5: Comparison of reconstruction methods on EndoNeRF

Comparison of reconstruction methods on EndoNeRF (surgical scenes), in terms of PSNR, SSIM, and LPIPS. The numbers of non-GS methods, FPS, and GPU usage (Mem.) are taken from . * denotes numbers taken from . denotes the average of the values reported in the original paper.

Method GS PSNR↑ SSIM↑ LPIPS↓ FPS↓ Mem.↓
EndoNeRF 36.06 0.93 0.09 0.04 19GB
EndoSurf 36.53 0.95 0.07 0.04 17GB
LerPlane-9k 35.00 0.93 0.08 0.91 20GB
LerPlane-32k 37.38 0.95 0.05 0.87 20GB
Endo-4DGS 37.00 0.96 0.05 - 4GB
EndoGaussian 37.85 0.96 0.05 195.09 2GB
HFGS 38.14 0.97 0.03 - -

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors