Focused Diffusion-GAN (FDGAN): generated Cityscapes-Pedestrian sample sets behind the reported results (access on request)

The 45 Cityscapes-Pedestrian sample sets (3,000 generated images each) behind the reported Cityscapes-Pedestrian results of the 11 models trained with the FDGAN code for 'Focused Diffusion-GAN: Object-Centric Image Generation Using Integrated GAN and Diffusion Frameworks' (Image and Vision Computing, under revision) (ADM, FDGAN* with the adversarial stage disabled, FDGAN and the ablation rows), packed as FDGAN_samples_cityscapes.tar: one zip per set, each holding the generated PNG images and a meta.json with the conditioning box of every image and the real training image it came from. The images are generated by models trained on Cityscapes crops, so access is granted on request to researchers holding a Cityscapes licence. Set names follow the job and sheet names of https://github.com/Latzi/focused-diffusion-gan. The checkpoints and the Traffic Signs and COCO potted plant sample sets are in the companion open record, DOI 10.5281/zenodo.23750546. SHA256SUMS.txt lists the SHA-256 checksum of every tar of both records.