nihaomiao / CVPR23_LFDM

The pytorch implementation of our CVPR 2023 paper "Conditional Image-to-Video Generation with Latent Flow Diffusion Models"
BSD 2-Clause "Simplified" License
443 stars 42 forks source link

What do the "deformed" and "prediction" mean? #39

Closed HARVEYDOU closed 10 months ago

HARVEYDOU commented 10 months ago

HI! 1700363922222 In codes: real_out_img_list.append(generated["prediction"]) real_warped_img_list.append(generated["deformed"]) 1700363554920 In the demo, what do the second and the third part seperately mean? As i know, they are probably "deformed" and "prediction". Could you describe it in detail. Appreciate for your patient answer

nihaomiao commented 10 months ago

"deformed" means directly applying the optical flow (deformation flow) to the original image to show the warped result, while "prediction" represents the final output image from the decoder of our network.

HARVEYDOU commented 10 months ago

Thanks a lot!

"deformed" means directly applying the optical flow (deformation flow) to the original image to show the warped result, while "prediction" represents the final output image from the decoder of our network.