Edges are not high quality when trained with Swin Tiny version

plemeri / InSPyReNet

Official PyTorch implementation of Revisiting Image Pyramid Structure for High Resolution Salient Object Detection (ACCV 2022)

MIT License

449 stars 69 forks source link

Edges are not high quality when trained with Swin Tiny version #18

Closed rakesh-reddy95 closed 1 year ago

rakesh-reddy95 commented 1 year ago

While training with SwinB (384 384) and keeping the dynamic resize to 1280 during inference I see a high quality edges without any lack of global object saliency. But when train a Tiny version of SwinTransformer (224 224) I see the global object saliency is lacking and not getting the high quality edges. May I know if I have to change anything for tiny versions or any suggestions here?

plemeri commented 1 year ago

Hi, try reducing dynamic resize into somewhere between 768 ~ 1024.