OptVQ (16x16x4)

Reported on 4 benchmarks across 1 task · 1 paper · 1 SOTA

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

Medical4 results

Image ReconstructiononImageNet
LPIPS· 2024-12-19
0.076
best: 0.066 (OptVQ (16x16x8))
SOTA
Preventing Local Pitfalls in Vector Quantization via Optimal Transport arXiv:2412.15195
Image ReconstructiononImageNet
FID· 2024-12-19
1
best: 0.49 (MGVQ (16x16x8))
Preventing Local Pitfalls in Vector Quantization via Optimal Transport arXiv:2412.15195
Image ReconstructiononImageNet
PSNR· 2024-12-19
26.59
best: 27.57 (OptVQ (16x16x8))
Preventing Local Pitfalls in Vector Quantization via Optimal Transport arXiv:2412.15195
Image ReconstructiononImageNet
SSIM· 2024-12-19
0.717
best: 0.787 (MGVQ (16x16x8))
Preventing Local Pitfalls in Vector Quantization via Optimal Transport arXiv:2412.15195