Check the preview of 2nd version of this platform
being developed by the open MLCommons taskforce on automation and reproducibility
as a free, open-source and technology-agnostic on-prem platform.
Not All Areas Are Equal: Transfer Learning for Semantic Segmentation via Hierarchical Region Selection
lib:346ca81365489b0b (v1.0.0)
Authors: Ruoqi Sun, Xinge Zhu, Chongruo Wu, Chen Huang, Jianping Shi, Lizhuang Ma
Where published: CVPR 2019 6
Document: PDF DOI
Abstract URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Not_All_Areas_Are_Equal_Transfer_Learning_for_Semantic_Segmentation_CVPR_2019_paper.html
The success of deep neural networks for semantic segmentation heavily relies on large-scale and well-labeled datasets, which are hard to collect in practice. Synthetic data offers an alternative to obtain ground-truth labels for free. However, models directly trained on synthetic data often struggle to generalize to real images. In this paper, we consider transfer learning for semantic segmentation that aims to mitigate the gap between abundant synthetic data (source domain) and limited real data (target domain). Unlike previous approaches that either learn mappings to target domain or finetune on target images, our proposed method jointly learn from real images and selectively from realistic pixels in synthetic images to adapt to the target domain. Our key idea is to have weighting networks to score how similar the synthetic pixels are to real ones, and learn such weighting at pixel-, region- and image-levels. We jointly learn these hierarchical weighting networks and segmentation network in an end-to-end manner. Extensive experiments demonstrate that our proposed approach significantly outperforms other existing baselines, and is applicable to scenarios with extremely limited real images.
Where published: CVPR 2019 6
Document: PDF DOI
Abstract URL: http://openaccess.thecvf.com/content_CVPR_2019/html/Sun_Not_All_Areas_Are_Equal_Transfer_Learning_for_Semantic_Segmentation_CVPR_2019_paper.html
The success of deep neural networks for semantic segmentation heavily relies on large-scale and well-labeled datasets, which are hard to collect in practice. Synthetic data offers an alternative to obtain ground-truth labels for free. However, models directly trained on synthetic data often struggle to generalize to real images. In this paper, we consider transfer learning for semantic segmentation that aims to mitigate the gap between abundant synthetic data (source domain) and limited real data (target domain). Unlike previous approaches that either learn mappings to target domain or finetune on target images, our proposed method jointly learn from real images and selectively from realistic pixels in synthetic images to adapt to the target domain. Our key idea is to have weighting networks to score how similar the synthetic pixels are to real ones, and learn such weighting at pixel-, region- and image-levels. We jointly learn these hierarchical weighting networks and segmentation network in an end-to-end manner. Extensive experiments demonstrate that our proposed approach significantly outperforms other existing baselines, and is applicable to scenarios with extremely limited real images.
Relevant initiatives
Related knowledge about this paper
Reproduced results (crowd-benchmarking and competitions)
Artifact and reproducibility checklists
- ACM/cTuning artifact appendix and reproducibility checklist
- NeurIPS reproducibility checklist
- SIGPLAN's checklist for empirical evaluation
- Collective Knowledge (organizing research projects based on FAIR principles)