Extract feature and key points using the demo

[tinaplus]

I want to extract the key points and corresponding feature vector using the demo. My question is the following, how can you do it using the notebook demo. I know perhaps keypoints are returned in the dictionary ( mkpts0 = batch['mkpts0_f'].cpu().numpy()
mkpts1 = batch['mkpts1_f'].cpu().numpy()).

How can I get the correponding feature vector. Perhaps in loftr.py these (feat_f0_unfold, feat_f1_unfold) are corresponding feature vectors ? could you please guide me.

I am someone who is new to the field, so excuse me if the questions are super obvious.