摘要
This paper proposes a novel end-to-end approach for salient object detection task to enhance the performances. The traditional downscaling method such as max pooling layer is replaced by detail preserving pooling layer to capture more effective features. Moreover, the squeeze and excitation block is adopted to extract image features with the channel wise importance. Finally, densely connected architecture is introduced to maximizes feature reuse and reduce the computational cost in the process of generate saliency maps. The proposed method, on several public benchmarks acquires competitive or better performances than other similar approaches.
源语言 | 英语 |
---|---|
主期刊名 | Proceedings of the 2019 2nd International Conference on Algorithms, Computing and Artificial Intelligence, ACAI 2019 |
出版商 | Association for Computing Machinery |
页 | 127-132 |
页数 | 6 |
ISBN(电子版) | 9781450372619 |
DOI | |
出版状态 | 已出版 - 20 12月 2019 |
活动 | 2nd International Conference on Algorithms, Computing and Artificial Intelligence, ACAI 2019 - Sanya, 中国 期限: 20 12月 2019 → 22 12月 2019 |
出版系列
姓名 | ACM International Conference Proceeding Series |
---|
会议
会议 | 2nd International Conference on Algorithms, Computing and Artificial Intelligence, ACAI 2019 |
---|---|
国家/地区 | 中国 |
市 | Sanya |
时期 | 20/12/19 → 22/12/19 |
指纹
探究 'Salient object detection with detail-preserving pooling and feature channel refinement' 的科研主题。它们共同构成独一无二的指纹。引用此
Yu, H., Chen, D., & Yi, H. (2019). Salient object detection with detail-preserving pooling and feature channel refinement. 在 Proceedings of the 2019 2nd International Conference on Algorithms, Computing and Artificial Intelligence, ACAI 2019 (页码 127-132). (ACM International Conference Proceeding Series). Association for Computing Machinery. https://doi.org/10.1145/3377713.3377725