search

oke-aditya / image_similarity

PyTorch Blog Post On Image Similarity Search
https://oke-aditya.github.io/image_similarity/
Apache License 2.0
249 stars 59 forks source link

Fine-tuning #4

Open fiastros opened 3 years ago

fiastros commented 3 years ago

I'm working on cat facial recognition. so my first idea was to calculate the distance between two embedded cat faces. Is it possible to retrain/fine tune the embedding models for such task ?

Thank you.

oke-aditya commented 3 years ago

Yes of coure you can retrain these models on any dataset. The training scripts are provided here. Note that this similarity search works on unlabelled data. It can help to compute distance between two cat faces.

But a few thoughts about cat facial recongition. You can probably label the data and train a object detector such as FasterRCNN. These are available in torchvision. It would work better and faster.

fiastros commented 3 years ago

Thank you for your advice. I have tried yolov5 for object detection: it works fine but the issue is that it needs lots of images for new faces, and it needs retrain each time....