search

suyukun666 / S2CNet

Official PyTorch implementation of the “Spatial-Semantic Collaborative Cropping for User Generated Content”. (AAAI24)
40 stars 4 forks source link

question about the paper and model #4

Open ppdk-data opened 8 months ago

ppdk-data commented 8 months ago

Hello, i have a question about the paper and model.

To generate the cropping result of the aspect ratio, it is necessary to create several candidate boxes by myself? The model's inputs is candidate boxes, and the outputs is the score?

ppdk-data commented 8 months ago

because the final output is mos, not the subimage. Is that correct?

suyukun666 commented 8 months ago

Hello, i have a question about the paper and model.

To generate the cropping result of the aspect ratio, it is necessary to create several candidate boxes by myself? The model's inputs is candidate boxes, and the outputs is the score?

Yes. you can refer to like：https://github.com/suyukun666/S2CNet/blob/main/data/croppingdataset.py#L204 generate the specific ratio boxes