How does RPN in Faster R-CNN work?
1 visualizzazione (ultimi 30 giorni)
Mostra commenti meno recenti
I would like to know how regionProposal layer can output correct (x,y,w,h).
![](https://www.mathworks.com/matlabcentral/answers/uploaded_files/730484/image.png)
rpnBoxDeltas layer is trained to output the correct (dx,dy,dw,dh).
rpnClassification is trained to output correct object or background classification degree too.
I appreciate it if someone answers my question. Thank you.
0 Commenti
Risposte (0)
Vedere anche
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!