Hi! I notice that the training process only needs the train_sharp videos in REDS dataset, the low-quality video inputs are automatically generated in the training pipeline. My question is how can I visualize the generated low-quality videos? I want to change the parameters of the quality degradation process, but I cannot see the degradation results.
Hi! I notice that the training process only needs the train_sharp videos in REDS dataset, the low-quality video inputs are automatically generated in the training pipeline. My question is how can I visualize the generated low-quality videos? I want to change the parameters of the quality degradation process, but I cannot see the degradation results.