tianrun-chen / SAM-Adapter-PyTorch

Adapting Meta AI's Segment Anything to Downstream Tasks with Adapters and Prompts
MIT License
850 stars 75 forks source link

FFT for High frequency component #44

Closed Abhiram-kandiyana closed 1 year ago

Abhiram-kandiyana commented 1 year ago

@tianrun-chen In the paper, It is mentioned that you used HFC of the image as part of the input image embedding. But I can't find the corresponding code in your Github repo. I was able to find the code in Explicit-Visual_prompt repo but not in yours and It looks like you have made some changes to their code to implement SAM. Did I miss something or have you not implemented FFT?

Abhiram-kandiyana commented 1 year ago

I found it. It is in https://github.com/tianrun-chen/SAM-Adapter-PyTorch/blob/main/models/mmseg/models/sam/image_encoder.py#L317C1-L318C1