feipan664:视频内容:
Cityscapes: Cityscapes testing demo video from https://www.cityscapes-dataset.com/;
Trained on GTA5: the segmentation network trained on GTA5 in a supervised way, without using any domain adaptation;
Trained on Cityscapes: the segmentation network trained on 2975 annotated images from Cityscapes dataset;
Ours (IntraDA): Our model uses GTA5 as source domain and unlabeled 2975 target images from Cityscapes.
feipan664:欢迎关注我们的CVPR2020的Oral工作:
Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision
Fei Pan, Inkyu Shin, Francois Rameau, Seokju Lee, In So Kweon KAIST, Korea
IEEE Conference on Computer Vision and Pattern Recoginition (CVPR), 2020 (Oral)
Github: https://github.com/feipan664/IntraDA.git
feipan664:Title: Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision
Cityscapes: Cityscapes testing demo video from https://www.cityscapes-dataset.com/;
Trained on GTA5: the segmentation network trained on GTA5 in a supervised way, without using any domain adaptation;
Trained on Cityscapes: the segmentation network trained on 2975 annotated images from Cityscapes dataset;
Ours (IntraDA): Our model uses GTA5 as source domain and unlabeled 2975 target images from Cityscapes.
Unsupervised Intra-domain Adaptation for Semantic Segmentation through Self-Supervision
Fei Pan, Inkyu Shin, Francois Rameau, Seokju Lee, In So Kweon KAIST, Korea
IEEE Conference on Computer Vision and Pattern Recoginition (CVPR), 2020 (Oral)
Github: https://github.com/feipan664/IntraDA.git