0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?

More than 1 year has passed since last update.

[論文備忘録] FPN (Feature Pyramid Networks for Object Detection), 2016

Posted at

Feature Pyramid Networks for Object Detection

ResNetなど、ConvNetでは強い特徴になるほど一般的には解像度が下がってしまう。
提案手法では、解像度が低く強い特徴マップから、解像度が高く強さも保った特徴マップを出力する。

Bottom-up pathway:
The bottom-up pathway is the feedforward computation of the backbone ConvNet.

Topdown pathway and lateral connections:
低解像度で強い特徴マップから高解像度な特徴マップを作っていく。
粗い特徴マップから最近傍補間で2倍にアップスケールしていく。
これらの特徴マップはbottom-up pathwayの同じサイズの特徴マップとマージされる。(lateral connection)
lateral connectionには1x1でチャネル数固定のconvolutional layerを掛ける。

FPN (Feature Pyramid Networks for Object Detection), 2016.png

FPN (Feature Pyramid Networks for Object Detection), 2016-1.png

FPN (Feature Pyramid Networks for Object Detection), 2016-2.png

0
0
0

Register as a new user and use Qiita more conveniently

  1. You get articles that match your needs
  2. You can efficiently read back useful information
  3. You can use dark theme
What you can do with signing up
0
0

Delete article

Deleted articles cannot be recovered.

Draft of this article would be also deleted.

Are you sure you want to delete this article?