czczup / ViT-Adapter

[ICLR 2023 Spotlight] Vision Transformer Adapter for Dense Predictions
https://arxiv.org/abs/2205.08534
Apache License 2.0
1.27k stars 140 forks source link

What is the principle of feature map visualization? #132

Closed facias914 closed 1 year ago

facias914 commented 1 year ago

Hi ! The research in this paper is an excellent work

By the way, I would like to ask what is the principle of feature map visualization in the paper. Is it simply a matter of regularizing each channel of the feature map and then adding up all the channels to take the mean?

Does the feature map visualization in the detection task have the same effect as the feature map visualization in the segmentation task?

czczup commented 1 year ago

Hi, please see the issue #76.