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】SsSMnet : self-superviseでtraining

Last updated at Posted at 2020-12-08

Self-Supervised Learning for Stereo Matching with Self-Improving Ability

![image.png](https://qiita-image-store.s3.ap-northeast-1.amazonaws.com/0/482094/b7a0ccab-0a69-e33b-0cac-81809ba1d283.png)

GC-NetをSelf-SuperviseでTraining出来るようにしたネットワーク。

新規性

Loss

今では一般的な方法ですが、Image warpingをする事で反対の画像をReconstructionして、Errorを計算する手法。

主に3つのLossがある

  1. LR Consistency Loss
    =>右と左のDisparityの違いを計算する
  2. Reconstruction Loss
    =>Reconstructした画像を入力画像と比べる。評価指標にSADやSSIMが良く使われる
  3. Disparity Smoothness Loss
    =>DisparityのSmoothさを計算する。Edgeに対するmaskをするとより良い

結論

・stereo画像を使う事でGround TruthなしでもDisparity(1/depth)のLossを計算する事が出来る。

参考文献

Self-Supervised Learning for Stereo Matching with Self-Improving Ability https://arxiv.org/pdf/1709.00930.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?