search

AngeLouCN / CaraNet

Context Axial Reverse Attention Network for Small Medical Objects Segmentation
462 stars 30 forks source link

do u do some ablation experments on ur modules #16

Closed wcyjerry closed 2 years ago

wcyjerry commented 2 years ago

Hi, thanks for ur great work, But I dont see ablation experiments whin ur net, like which part effective the performence more , Do u have done it? I'm very interesting in which modules matters most. thx

AngeLouCN commented 2 years ago

Hi, thank you for your interest. We do not do ablation study in this paper. But you can compare our architecture and performance with PraNet. Also, some recent citations of this paper are discussing details about the CFP and A-RA module, may be it will be helpful.

wcyjerry commented 2 years ago

could u please share the names of paper compare about it pls thx

------------------ Original ------------------ From: AngeLouCN @.> Date: Fri,May 27,2022 10:49 AM To: AngeLouCN/CaraNet @.> Cc: q671383789 @.>, Author @.> Subject: Re: [AngeLouCN/CaraNet] do u do some ablation experments on ur modules (Issue #16)

Hi, thank you for your interest. We do not do ablation study in this paper. But you can compare our architecture and performance with PraNet. Also, some recent citations of this paper are discussing details about the CFP and A-RA module, may be it will be helpful.

— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>

AngeLouCN commented 2 years ago

Hi, you can check those paper. An Attention-Preserving Network-Based Method for Assisted Segmentation of Osteosarcoma MRI Images ColonFormer: An Efficient Transformer based Method for Colon Polyp Segmentation Attention based multi-scale parallel network for polyp segmentation