Lei

The output offset values are very small (like 0.01, -0.0003). What exactly do these values mean? Are these pixel-level values?

> Hi, have u figured it out? No I didn't do it

I still don't know what gate_activation should be

@hadonga Didn't get it. I was using the code directly. So 'ngpu' is relatted to 'gate_activation' ?

> We have released the API for FastSAM. Please see for details at fastsam/decoder.py. > > import model > > ```python > from fastsam import FastSAM ,FastSAMDecoder > model =...

Hi, I have the exactly same error.

> I don't why, but I changed the code in utils.py 17-20, then it works well. > > # Adjust boxes > ``` > boxes[:, 0] = torch.where(boxes[:, 0] <...

Yes. mIoU is just the mean of the positive IoU and the negative IoU.

Thanks for following and Sorry for the delay. The training & evaluation codes to the BLU dataset have been released. The data is also available.

Hi, There are two versions of the segmentation network. ED_FCN is a basic encoder-decoder FCN, while FCN_SR is the described FCN with shape refinement modules. I do not think there...