ICThruU Thank you for trying out Tonfotos
- If you just started indexing, application may not have yet enough statistics to do good time prediction. Give it a little bit more time (at least hour of indexing), then it’s prediction will become more accurate. Also, you may want to disable indexing of video for now, as video processing is very time consuming. You will be able to re-enable it later on, when you finish with your photos. Also, if you have CUDA-compatible GPU, you can install CUDA-Plugin and this will help you to reduce indexing time several times due to faster face recognition. Overall, processing speed depends on many factors, and the most important is drive reading speed. If you work on low speed network connection with a shared drive, that can really take quite long. Secong factor - is your processing capabilities (CPU speed and availability of GPU)
- All recognition results will be in your database already, so you will not need to index once again. You will just be able to add more people to your database, that’s it.
- Application does not rescan images unless you modify files, then only modified ones get processed once again.
- Yes, application is learning on your input, that is the whole idea. It’s suggestion can improve based on your input, but if you do mistakes this can actually make things worse. However, you can always go back and correct your mistakes. And no, there is no guarantee that program will stop doing mistakes. Mistakes are inevitable, unfortunately, more about that here: https://community.tonfotos.com/d/228-people-tagging-logic-seems-to-have-gone-astray/3 Aslo, as soon as applications runs out of “good suggestions” it will start offering more questionable ones. But this is not because program does not learn, it is because no more good ones left.
