Have a problem in Feature2D

In opencv’s feature2D, many useful feature extraction methods are provided, and we can easily extract, describe, and match the features. But if we use feature extraction methods not included in opencv for point extraction, how should we process the extracted points in this column so that we can continue to use opencv for description and matching tasks?

For example, I use the open source feature point extraction algorithm (GitHub - bluedream1121/Key.Net_PyTorch: [Reproduce] KeyNet (Barroso-Laguna et al. ICCV 2019) implementation using PyTorch. (Unofficial)) for feature point extraction, and after the extraction is done I want to modify it with sift descriptor and match it with flann, how should I implement it?

I guess I may have to convert the format of the feature points extracted by Key.Net_PyTorch to be suitable for the next operation, how should I convert it? Is there any code that I can refer to? Thank you very much for taking your time to read my question, and I sincerely hope to get your advice!