Re: [Shotwell] Development environment setup or "How to build shotwell from source?"




Thanks, compiled and run-tested on Debian/sid.


Face detection is - well - mediocre at best.

The pure detection is a bit weird. Sometimes it works perfect on the
same image it doesn't work on at all the day before. I have not yet
found out why that should be.

Face recognition as far as I see not existent.
I guess I have to test more :-(

For the recognition to work, you need to download the two extra files
mentioned in the flatpak manifest:

https://raw.githubusercontent.com/opencv/opencv_3rdparty/dnn_samples_fa
ce_detector_20180205_fp16/res10_300x300_ssd_iter_140000_fp16.caffemodel

https://raw.githubusercontent.com/opencv/opencv/master/samples/dnn/face_detector/deploy.prototxt

Sorry, I forgot in which folder they go. I think
~/.local/share/shotwell. That should also improve the detection itself
a bit.

Then again, I also haven't done a real test with OpenCV 4




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]