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 3 years have passed since last update.

【Stereo Depth】CRL(Cascade Residual Learning) : 2Stage Depth Estimation

Posted at

Cascade Residual Learning: A Two-stage Convolutional Neural Network for Stereo Matching

![image.png](https://qiita-image-store.s3.ap-northeast-1.amazonaws.com/0/482094/1a0e2e06-4d83-aa75-c740-c746828ea548.png)

どうやってネットワークを早くしようかと考えていたら、2stageにするが良いという結論になったそうです。

新規性

DispFullNet

従来研究と同じで、Correlation(相関性)を見るためにInner Product(内積)でMatching Scoreを計算し、いくつかの解像度でDisparityを推定し、それぞれの解像度でlossを計算している。

DispResNet

DispFullNetで得たDisparity MapからLeft Image ReconstructionしてEorrorも計算して、それらを入力にしてるのが図からわかりますね。

図がごちゃごちゃしててわかりにくいですが、ようはReconstruction Errorを使ってDisparityをRefinementしようということ。

結果

![image.png](https://qiita-image-store.s3.ap-northeast-1.amazonaws.com/0/482094/72969d32-9f6a-78c7-a57a-8d318ba3f22c.png) GC-Netより早い!

結論

・1stageより2stageの方が効率的だったりもする。

参考文献

Cascade Residual Learning: A Two-stage Convolutional Neural Network for Stereo Matching https://arxiv.org/pdf/1708.09204.pdf
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?