hongfz16 / DS-Net

[CVPR 2021/TPAMI 2023] Rank 1st in the public leaderboard of SemanticKITTI Panoptic Segmentation (2020-11-16)
MIT License
243 stars 29 forks source link

About Consensus-driven Fusion module #13

Closed DayDayUpUpUp666 closed 2 years ago

DayDayUpUpUp666 commented 2 years ago

Hi, thanks for your great work. I wonder to know that how the Consensus-driven Fusion module implementing?I found the merge_ins_sem in common_utils. But I donot find how to use the 'merge_ins_sem' in other code.

hongfz16 commented 2 years ago

Hi. Thank you for your interest. Yes the consensus-driven fusion is implemented in the merge_ins_sem function in common_utils. And it is called in this line: https://github.com/hongfz16/DS-Net/blob/72c3c251dd57074c3ce30a489fb47710a9fb5c1a/network/model_zoo.py#L348.

If you wish to use this module in your code, I think it is quite easy to implement it yourself. You travel through all segmented instances. For each one of them, you perform a majority voting in terms of semantic labels for all the points of the instance. The idea is to unify the semantic labels inside each instance. If you have further questions, feel free to reopen it or add another post.