Closed Foxerity closed 8 months ago
I mean the grid in the first row of the picture below. I know the lower row is because it is not trained well.
Hi, @Foxerity, have you loaded the correct pre-trained stage-one latent autoencoder? The ground truth flow should not have any issues.
the grid generated well in the demo file
the grid generated in the train_video_flow_diffusion_mhad_multiGPU file
I am referring to the GT grid. The grid in the demo is square, but in the file below is folded. I'm using the same AE weights.