search

facebookresearch / adaptive_teacher

This repo provides the source code for "Cross-Domain Adaptive Teacher for Object Detection".
Other
183 stars 36 forks source link

Question about Figure 4 #4

Closed tmp12316 closed 2 years ago

tmp12316 commented 2 years ago

Thank you for your work. But I have a question about Figure 4 in the main paper.

It seems that with only 10k iterations of the source-only pre-training, the model has achieved around 33.0 mAP, which has significantly outperformed the well-trained source-only results (28.8). Does it mean that the detectron2 implemented FRCNN works better?

yujheli commented 2 years ago

The result of high performance reported in the first iteration was after the first update of EMA (with one iteration update using pseudo labels). Though, we also think implementation with detectron2 usually has higher performance.

tmp12316 commented 2 years ago

That's okay, thanks.