Saliency map of SOD and COD

NiFangBaAGe / Explicit-Visual-Prompt

[CVPR 2023] Explicit Visual Prompting for Low-Level Structure Segmentations

https://nifangbaage.github.io/Explicit-Visual-Prompt/

BSD 3-Clause "New" or "Revised" License

174 stars 12 forks source link

Saliency map of SOD and COD #14

Open Sssssuperior opened 9 months ago

Sssssuperior commented 9 months ago

Hello, this is a very good job. We recently want to compare our results with EVP, but the provided results do not include SOD saliency maps. Also, when I load the EVP pretrained model, it prompts that the main component is "freq_adaptor", but this seems to be FP's parameter? Can you please tell me if I made any mistakes here? Or can you provide your saliency maps? Thank you!

NiFangBaAGe commented 9 months ago

Thanks for your interest in our work. 'freq_adaptor' in an extended version here, please specify model['encoder_mode']['name'] to 'fp' in the config file to use it, and find the corresponding model here. If you want to compare your results with EVP, please specify model['encoder_mode']['name'] to 'evp' and use the corresponding model.