search

LiWentomng / boxlevelset

The code for "Box-supervised Instance Segmentation with Level Set Evolution(ECCV2022)"
Apache License 2.0
185 stars 25 forks source link

No length regularization in LevelsetLoss? #15

Closed Yangr116 closed 1 year ago

Yangr116 commented 1 year ago

Hi, why is there not the third term (length regularization) in LevelsetLoss as described in Eq.3 of the paper? How many gains/drops will it bring?

LiWentomng commented 1 year ago

@Yangr116 Hi, in the ablation study, the length regularization of Level set loss is adopted, which is beneficial for the performance (especially when the performance is low). In the main results, I found that the length regularization can bring the 0.2~0.4mAP drop. So, I commented the calculation for the better final results.

Yangr116 commented 1 year ago

Thanks for your reply!