Gengzigang comments

Results 17 comments of


                                            Gengzigang

Increase length of Grade Value Qualifier

Thank you for your interest. Please refer to Section 4.9 in our paper(https://arxiv.org/pdf/2309.03895.pdf) for more information.

Will the proposed datasets be released ?

Thank you for your interest in the proposed datasets. Due to some copyright concerns, we currently have no plans to release these specific datasets. However, we believe that the publicly...

Will the proposed datasets be released ?

If you would like to train on all tasks (955k images per epoch) simultaneously, it took us 3.5 days to train for 200 epochs using 48 V100 GPUs. However, if...

Questions about the post-processing for keypoint detection and segmentation

Extracting keypoints or masks from the images generated by our editor is a simple mapping process, so the UNet will have minimal impact on performance.

Is there a printing error in the formula of the paper?

Thanks for your interest. You can check this folder for instructions for each task: https://github.com/cientgu/InstructDiffusion/tree/main/dataset/prompt.

Is there a printing error in the formula of the paper?

For editing, we use the instructions provided within the dataset itself. During the inference process, you are free to be creative.

Can the code generate 19 dimension heatmaps?

You should modify the code of data processing, the dimensions of the predictors of the model.