Image Generation On Cityscapes 25K 256X512

FID

Results

Performance results of various models on this benchmark

Comparison Table

Model Name	FID
semantic-bottleneck-scene-generation	62.97