Few-Shot Semantic Segmentation based on Detail-Preserving-Aware Loss

Chih Chung Hsu, Sin Di Ma

研究成果: Conference contribution

摘要

The vision-based autonomous driving techniques have been activated recently. The pixel-level prediction, semantic segmentation, is widely used in autonomous driving to acquire the pixel-level semantic annotation for further decision-making. However, traditional semantic segmentation requires a large training set to obtain promising performance. It is essential that it is hard to collect such a large dataset for every scenario for autonomous driving, such as outdoor or other bad weather scenarios. In this paper, we propose a novel boundary-aware loss incorporating the rare object augmentation techniques to boost the performance under a limited training set. The conventional edge extraction operator is applied in the ground truth to obtain the boundary information, as well as the detailed branch is proposed to approximate the predicted results to have a similar boundary with the ground truth. Such hard constraints result in the network being hard to overfit, as well as improve the robustness of the semantic segmentation. Extensive experiments demonstrated the proposed method's effectiveness, especially in limited training set scenarios.

原文English
主出版物標題Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
發行者Institute of Electrical and Electronics Engineers Inc.
頁面581-582
頁數2
ISBN(電子)9781665470506
DOIs
出版狀態Published - 2022
事件2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022 - Taipei, Taiwan
持續時間: 2022 7月 62022 7月 8

出版系列

名字Proceedings - 2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022

Conference

Conference2022 IEEE International Conference on Consumer Electronics - Taiwan, ICCE-Taiwan 2022
國家/地區Taiwan
城市Taipei
期間22-07-0622-07-08

All Science Journal Classification (ASJC) codes

  • 人工智慧
  • 電腦科學應用
  • 硬體和架構
  • 可再生能源、永續發展與環境
  • 電氣與電子工程
  • 媒體技術
  • 健康資訊學
  • 儀器

指紋

深入研究「Few-Shot Semantic Segmentation based on Detail-Preserving-Aware Loss」主題。共同形成了獨特的指紋。

引用此