Mask Generation with Meta-Learning Classifier Weight Transformer Network for Few-Shot Image Segmentation

Fong Ci Jhou, Kai Wen Liang, Chung Hsun Lo, Chien Yao Wang, Yung Fang Chen, Jia Ching Wang, Pao Chi Chang

研究成果: 書貢獻/報告類型會議論文篇章同行評審

摘要

This paper proposes a meta-learning classification weight transfer network to generate masks as a few-shot image segmentation architecture. It generates good prior masks via a pretrained classification weight transfer architecture, and uses a pretrained feature extraction architecture on query images and support images. The network architecture exploits a top-down path in a feature augmentation module to adaptively transfer information from fine to coarse features for extracting features from query images. Finally, the classification module predicts the segmentation of the query image. The experimental results show that using the mean intersection of joints (mIOU) as the evaluation mechanism, the accuracy of the 1-shot experimental results is 1.7% higher than that of the baseline. In the 5-shot experimental results, the accuracy is also improved by 2.6%. Therefore, compared with the baseline, it clearly shows that the mask generated by the meta-learning classification weight transfer network can effectively help improve the performance of few-shot image segmentation system.

原文???core.languages.en_GB???
主出版物標題2023 International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2023 - Proceedings
發行者Institute of Electrical and Electronics Engineers Inc.
頁面457-458
頁數2
ISBN(電子)9798350324174
DOIs
出版狀態已出版 - 2023
事件2023 International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2023 - Pingtung, Taiwan
持續時間: 17 7月 202319 7月 2023

出版系列

名字2023 International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2023 - Proceedings

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???2023 International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2023
國家/地區Taiwan
城市Pingtung
期間17/07/2319/07/23

指紋

深入研究「Mask Generation with Meta-Learning Classifier Weight Transformer Network for Few-Shot Image Segmentation」主題。共同形成了獨特的指紋。

引用此