Decoupled-Classification-Refinement icon indicating copy to clipboard operation
Decoupled-Classification-Refinement copied to clipboard
verified publisher icon SHI-Labs

Question about technical details

Open hdjang opened this issue 6 years ago • 1 comments

Hi, thanks for sharing your great work.

I have a question regarding technical details of your works.

  1. In Table.1 in your paper(ECCV2018), you compare 'sampling methods' to choose training samples of DCR module. Can you let me know 'precise' pool of samples in each sampling method? It looks a little confusing for me since 'FP only (hard false positive only)' can also include foreground(FG) or background(BG) samples. Did you mean the sampling pool of each sampling method as pool that 'intersects' all cases(e.g., FP, FG, and BG)?

hdjang avatar Mar 18 '19 09:03 hdjang

Please refer to these codes: https://github.com/bowenc0221/Decoupled-Classification-Refinement/blob/dcr_v1/dcr_v1/core/rcnn.py#L73-L93 for the exact sampling method.

bowenc0221 avatar Jul 29 '19 04:07 bowenc0221