paper-with-me

홈 › Papers

1st Place Solution to VisDA-2020: Bias Elimination for Domain Adaptive Pedestrian Re-identification

2020-12-25 · Jianyang Gu, Hao Luo, Weihua Chen, Yiqi Jiang, Yuqi Zhang, Shuting He, Fan Wang, Hao Li, Wei Jiang

This paper presents our proposed methods for domain adaptive pedestrian re-identification (Re-ID) task in Visual Domain Adaptation Challenge (VisDA-2020). Considering the large gap between the source domain and target domain, we focused on solving two biases that influenced the performance on domain adaptive pedestrian Re-ID and proposed a two-stage training procedure. At the first stage, a baseline model is trained with images transferred from source domain to target domain and from single camera to multiple camera styles. Then we introduced a domain adaptation framework to train the model on source data and target data simultaneously. Different pseudo label generation strategies are adopted to continuously improve the discriminative ability of the model. Finally, with multiple models ensembled and additional post processing approaches adopted, our methods achieve 76.56% mAP and 84.25% rank-1 on the test set. Codes are available at https://github.com/vimar-gu/Bias-Eliminate-DA-ReID

📄 PDF Abstract BibTeX arXiv:2012.13498

Code (1)

vimar-gu/Bias-Eliminate-DA-ReID 공식 구현 pytorch

Tasks

Domain AdaptationPseudo Label

Similar Papers 제목 키워드 기반

MixMatch Domain Adaptaion: Prize-winning solution for both tracks of VisDA 2019 challenge

2019-10-09 · Danila Rukhovich, Danil Galeev

We present a domain adaptation (DA) system that can be used in multi-source and semi-supervised settings. Using the proposed method we achieved 2nd place on multi-source track and 3rd place on semi-supervised track of th…

Domain Adaptation

1st Place Solution to NeurIPS 2022 Challenge on Visual Domain Adaptation

2022-11-26 · Daehan Kim, Minseok Seo, YoungJin Jeon, Dong-Geol Choi

The Visual Domain Adaptation(VisDA) 2022 Challenge calls for an unsupervised domain adaptive model in semantic segmentation tasks for industrial waste sorting. In this paper, we introduce the SIA_Adapt method, which inco…

Domain AdaptationSemantic Segmentation

2nd Place Solution for VisDA 2021 Challenge -- Universally Domain Adaptive Image Recognition

2021-10-27 · Haojin Liao, Xiaolin Song, Sicheng Zhao, Shanghang Zhang 외

The Visual Domain Adaptation (VisDA) 2021 Challenge calls for unsupervised domain adaptation (UDA) methods that can deal with both input distribution shift and label set variance between the source and target domains. In…

Domain AdaptationUniversal Domain AdaptationUnsupervised Domain Adaptation

VisDA: The Visual Domain Adaptation Challenge

2017-10-18 · Xingchao Peng, Ben Usman, Neela Kaushik, Judy Hoffman 외

We present the 2017 Visual Domain Adaptation (VisDA) dataset and challenge, a large-scale testbed for unsupervised domain adaptation across visual domains. Unsupervised domain adaptation aims to solve the real-world prob…

Domain AdaptationGeneral Classificationimage-classificationImage Classification+3

Pre-Training Transformers for Domain Adaptation

2021-12-18 · Burhan Ul Tayyab, Nicholas Chua

The Visual Domain Adaptation Challenge 2021 called for unsupervised domain adaptation methods that could improve the performance of models by transferring the knowledge obtained from source datasets to out-of-distributio…

Domain AdaptationUnsupervised Domain Adaptation