3D Human Reconstruction On 4D Dress

Metrics

Chamfer (cm)

IoU

Normal Consistency

Results

Performance results of various models on this benchmark

Model Name	Chamfer (cm)	IoU	Normal Consistency	Paper Title
SelfRecon_Inner	3.180	0.754	0.729	SelfRecon: Self Reconstruction Your Digital Avatar from Monocular Video
SNARF_Inner	1.158	0.907	0.843	SNARF: Differentiable Forward Skinning for Animating Non-Rigid Neural Implicit Shapes
PaMIR_Inner	2.520	0.706	0.805	PaMIR: Parametric Model-Conditioned Implicit Representation for Image-based Human Reconstruction
PIFu_Inner	2.696	0.690	0.792	PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization
PIFuHD_Outer	2.393	0.743	0.763	PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization
PaMIR_Outer	2.608	0.715	0.777	PaMIR: Parametric Model-Conditioned Implicit Representation for Image-based Human Reconstruction
SCANimate_Outer	1.237	0.912	0.828	SCANimate: Weakly Supervised Learning of Skinned Clothed Avatar Networks
ECON_Outer	2.852	0.728	0.760	ECON: Explicit Clothed humans Optimized via Normal integration
Vid2Avatar_Outer	4.027	0.745	0.683	Vid2Avatar: 3D Avatar Reconstruction from Videos in the Wild via Self-supervised Scene Decomposition
SNARF_Outer	1.248	0.930	0.827	SNARF: Differentiable Forward Skinning for Animating Non-Rigid Neural Implicit Shapes
ICON_Outer	2.832	0.756	0.762	ICON: Implicit Clothed humans Obtained from Normals
ECON_Inner	2.543	0.736	0.796	ECON: Explicit Clothed humans Optimized via Normal integration
SiTH_Inner	2.110	0.755	0.824	SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion
ICON_Inner	2.473	0.752	0.798	ICON: Implicit Clothed humans Obtained from Normals
SelfRecon_Outer	3.014	0.787	0.725	SelfRecon: Self Reconstruction Your Digital Avatar from Monocular Video
PIFu_Outer	2.783	0.697	0.759	PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization
X-Avatar_Outer	1.177	0.946	0.841	X-Avatar: Expressive Human Avatars
SiTH_Outer	2.322	0.749	0.794	SiTH: Single-view Textured Human Reconstruction with Image-Conditioned Diffusion
PIFuHD_Inner	2.426	0.739	0.793	PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization
SCANimate_Inner	0.965	0.918	0.854	SCANimate: Weakly Supervised Learning of Skinned Clothed Avatar Networks

0 of 22 row(s) selected.