-
## 一言でいうと
各クラス1サンプルのみの分類を半教師で行う手法(One-shot半教師)の精度を高めた研究。教師あり+教師なしのlossを組み合わせており後者はAugmentationの強弱で予測不変(Consistency)と表現一致(Contrastive)を使用。学習では確信度高いサンプルを追加するBootstrappingを使用
### 論文リンク
https://a…
-
1.TexturePose: Supervising Human Mesh Estimation with Texture Consistency(2019)
Texture map (texel): A corresponding UVmap un-warps the template surface onto an image, A, which is the texture map
co…
-
- https://arxiv.org/abs/2106.09018
- 2021 ICCV
本論文では、これまでの複雑なマルチステージ手法とは対照的に、エンド・ツー・エンドの半教師付きオブジェクト検出アプローチを紹介する。
エンド・ツー・エンドの学習により、カリキュラム中に疑似ラベルの品質が徐々に向上し、より正確な疑似ラベルが得られることで、物体検出の学習に役立つ。
また、このフレ…
e4exp updated
3 years ago
-
It seems that in `/semi_supervised_segmentation/losses.py` line 195, it should be `ce_loss = self.ce(net_output, target.long()) if self.weight_ce != 0 else 0` instead of `ce_loss = self.ce(net_output,…
-
The process of reproducing the code has the following questions：
1. Is this code using supervised learning, semi-supervised learning or unsupervised learning?
2. Where is your training model and tra…
-
%% One-class SVM
% OCC - SIFT
[SVMModel1, score1, acc1_max, boundary1] = trainSVM(SIFT_trainOCC, labels_trainOCCNum, SIFT_test, labels_testNum, 0.04, 1);
% OCC - SURF
[SVMModel2, score2, acc2_…
-
For example, when using KMeans, input some data that already has labels to assist in clustering or initialize the `cluster_centers_` ?
Thanks and look forward to your reply.
-
Wow! You did a great analysis pipeline which i want to do. Thank you for your great job.
And, i want to discuss some questions about metacell analysis as following:
-
Hi,
Thanks for your sharing.
I try to implement the auxiliary semi supervised model https://arxiv.org/pdf/1602.05473.pdf using kaos and adapting your code from the semi supervised model, but I fail…
-
Hi, thanks for sharing excellent work on MICCAI 2020!
When trying to implement semi-supervised segmentation task, because no Semi-supervised segmentation (./semi-seg/) available, I change directory t…