Conditional Image Generation On Imagenet 1
Metriken
FID
Inception score
Ergebnisse
Leistungsergebnisse verschiedener Modelle zu diesem Benchmark
Vergleichstabelle
| Modellname | FID | Inception score |
|---|---|---|
| instance-selection-for-gans | 9.07 | 37.1 |
| feature-quantization-improves-gan-training | 9.67 | 25.96 |
| instance-conditioned-gan | 6.7 | 45.9±0.3 |
| instance-conditioned-gan | 10.2±0.1 | 30.1±0.1 |