NVlabs / AL-MDN

Official pytorch implementation of Active Learning for deep object detection via probabilistic modeling (ICCV 2021)
https://openaccess.thecvf.com/content/ICCV2021/html/Choi_Active_Learning_for_Deep_Object_Detection_via_Probabilistic_Modeling_ICCV_2021_paper.html
Other
168 stars 23 forks source link

How to use GMM in Faster RCNN? #11

Closed ChuQiaosong closed 2 years ago

ChuQiaosong commented 2 years ago

Thanks for your great work! As you have mentioned in your paper, GMM works well on Faster RCNN. So can you share your code on Faster RCNN? I want to do some experiments on a two-stage detector.

jwchoi384 commented 2 years ago

Hi @ChuQiaosong, Sorry for the late reply. I do not currently have the implemented Faster RCNN code. However, the implementation is quite simple. You can set additional heads in the output layer of the Faster RCNN and implement the proposed loss functions on Faster RCNN's loss part by referring to this repo.