摘要
This paper introduces Permutation Transient Attention Encoders as a novel and efficient approach for NLOS reconstruction. Our method, utilizing Transient Attention Blocks and Permute-MLP, enhances transient feature extraction and offers more discriminative feature representations than existing approaches.
源语言 | 英语 |
---|---|
DOI | |
出版状态 | 已出版 - 2024 |
活动 | Computational Optical Sensing and Imaging, COSI 2024 - Part of Optica Imaging Congress - Toulouse, 法国 期限: 15 7月 2024 → 19 7月 2024 |
会议
会议 | Computational Optical Sensing and Imaging, COSI 2024 - Part of Optica Imaging Congress |
---|---|
国家/地区 | 法国 |
市 | Toulouse |
时期 | 15/07/24 → 19/07/24 |