search

aim-uofa / AdelaiDet

AdelaiDet is an open source toolbox for multiple instance-level detection and recognition tasks.
https://git.io/AdelaiDet
Other
3.37k stars 647 forks source link

Question about the semantic segmentation loss in CondInst #19

Open Scalsol opened 4 years ago

Scalsol commented 4 years ago

Thanks for your great work! I notice that adding an additional semantic segmentation loss in CondInst could boost the overall performance by about 1 mAP, which is quite a promising result! I want to ask:

  1. where do you add the semantic segmentation loss? After P3 with a 1x1 conv (just like YOLACT) or after F_{mask} or anywhere else?
  2. how do you calculate the semantic segmentation loss? Like YOLACT with sigmoid over 80 classes, or like HTC with 183 classes?

Waiting for your reply!

tianzhi0549 commented 4 years ago

@Scalsol Thank you for your questions. 1) The semantic segmentation task is after P_3 with a few 3x3 conv. layers. 2) We use an 80-class focal loss here.

Scalsol commented 4 years ago

I see. Thanks for your prompt reply :)