ML-PIFu (alternate)

Reported on 6 benchmarks across 2 tasks · 1 paper · 2 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision6 results

Object ReconstructiononBUFF
Point-to-surface distance (cm)· 2020-04-01
1.63
SOTA
PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization arXiv:2004.00452
3D Object ReconstructiononBUFF
Point-to-surface distance (cm)· 2020-04-01
1.63
SOTA
PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization arXiv:2004.00452
Object ReconstructiononBUFF
Chamfer (cm)· 2020-04-01
1.73
best: 0.61 (ARCH++)
PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization arXiv:2004.00452
Object ReconstructiononBUFF
Surface normal consistency· 2020-04-01
0.133
best: 0.22 (ML-PIFu (end-to-end))
PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization arXiv:2004.00452
3D Object ReconstructiononBUFF
Chamfer (cm)· 2020-04-01
1.73
best: 0.61 (ARCH++)
PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization arXiv:2004.00452
3D Object ReconstructiononBUFF
Surface normal consistency· 2020-04-01
0.133
best: 0.22 (ML-PIFu (end-to-end))
PIFuHD: Multi-Level Pixel-Aligned Implicit Function for High-Resolution 3D Human Digitization arXiv:2004.00452