[2011.00147] Pixel-Level Cycle Association: A New Perspective for Domain Adaptive Semantic Segmentationopen searchopen navigation menucontact arXivsubscribe to arXiv mailings

Domain adaptive semantic segmentation aims to train a model performing satisfactory pixel-level predictions on the target with only out-of-domain (source) annotations. The conventional solution to this task is to minimize the discrepancy between source and target to enable effective knowledge transfer. Previous domain discrepancy minimization methods are mainly based on the adversarial training. They tend to consider the domain discrepancy globally, which ignore the pixel-wise relationships and are less discriminative. In this paper, we propose to build the pixel-level cycle association between source and target pixel pairs and contrastively strengthen their connections to diminish the domain gap and make the features more discriminative. To the best of our knowledge, this is a new perspective for tackling such a challenging task. Experiment results on two representative domain adaptation benchmarks, i.e. GTAV $\rightarrow$ Cityscapes and SYNTHIA $\rightarrow$ Cityscapes, verify the ef

2 mentions: @AkiraTOSEI@AkiraTOSEI
Date:

Referring Tweets

@AkiraTOSEI
@AkiraTOSEI t.co/yVX8HAoCYA 領域分割タスクにおいて、ランダム抽出したsourceパッチ→その最近傍targetパッチ→その最近傍sourceパッチのラベルが同じものになるように制約をかけることで、ドメイン適応を行う研究。先行研究を大きく超える結果。 t.co/6Vl5Cju4PF
@AkiraTOSEI
@AkiraTOSEI t.co/yVX8HAoCYA A study of domain adaptation in semantic segmentation task by constraining the labels of randomly extracted source patches >its nearest neighbor target patches >its nearest neighbor source patches to be the same. The results greatly exceed previous results t.co/cnjBOVAkmx

Related Entries

[2006.10337] Category-Specific CNN for Visual-aware CTR Prediction at JD.comopen searchopen navigati...
Read more [2006.10337] Category-Specific CNN for Visual-aware CTR Prediction at JD.comopen searchopen navigati...
0 users, 2 mentions 2020/11/12 12:59
[2012.00364] Pre-Trained Image Processing Transformer
Read more [2012.00364] Pre-Trained Image Processing Transformer
0 users, 2 mentions 2020/12/23 11:21
[2012.06044] Mesoscopic photogrammetry with an unstabilized phone camera
Read more [2012.06044] Mesoscopic photogrammetry with an unstabilized phone camera
0 users, 2 mentions 2020/12/25 17:21
[2010.04456] Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting
Read more [2010.04456] Augmenting Physical Models with Deep Networks for Complex Dynamics Forecasting
0 users, 2 mentions 2021/02/12 11:21
MAE/SimMIM for Pre-Training Like a Masked Language Model | by Akihiro FUJII | Dec, 2021 | Medium
Read more MAE/SimMIM for Pre-Training Like a Masked Language Model | by Akihiro FUJII | Dec, 2021 | Medium
0 users, 1 mentions 2021/12/23 22:09

ML-Newsについて

機械学習の技術に関する情報は流速も早いし、分野も多様でキャッチアップが大変です。Twitterで機械学習用のリストを作っても、普段は機械学習以外の話題が多く流れており、効率的に情報収集するのは困難です。

ML-NewsはSNSを情報源とした機械学習に特化したニュースサイトです。機械学習に関する論文ブログライブラリコンペティション発表資料勉強会などの最新の情報を効率的に収集できます。

機械学習を応用した自然言語処理、画像認識、情報検索などの分野の情報や機械学習で必要になるデータ基盤やMLOpsの話題もカバーしています。
安定したサイト運営のためにGitHub sponsorを募集しています。

お知らせ

  • 2021/12/31: デザインを刷新しました
  • 2021/04/08: 日本語Kaggleのカテゴリを新設しました