ASIS
收藏资源简介:
发布了一个恶劣场景图像拼接数据集(ASIS),该数据集整合了低光照、雾霾和水下环境下的图像。ASIS包含2,250对图像,包括每种环境(低光照、水下和雾霾)各750张图像,涵盖了洞穴、沉船和田野等17个场景。这些数据部分来自互联网,部分来自独立拍摄。拍摄的图像远离平面结构,确保了所提出的ASIS数据集的视差多样性。
We present a challenging scene image stitching dataset named ASIS, which integrates images captured under low-light, haze, and underwater environments. ASIS comprises 2,250 image pairs, with 750 pairs corresponding to each of the three environments, covering 17 scenarios including caves, shipwrecks, and fields. A portion of the dataset is sourced from the Internet, while the remaining part was captured independently. All captured images avoid planar structures, thus ensuring the parallax diversity of the proposed ASIS dataset.
ACDIS 数据集概述
数据集名称
ASIS(Adverse Scene Image Stitching Dataset)
发布信息
- 发布时间:2025年10月12日
- 发布机构:相关论文发表于NeurIPS 2025
数据集内容
- 数据规模:包含2,250对图像
- 环境类型:
- 低光照环境:750张图像
- 水下环境:750张图像
- 雾霾环境:750张图像
- 场景覆盖:涵盖17种不同场景(包括洞穴、残骸、田野等)
数据特征
- 数据来源:部分来自互联网,部分为独立拍摄
- 结构特性:拍摄图像远离平面结构,确保数据集的视差多样性
- 标注信息:提供单应性参考,包含端点偏移量
获取方式
- Google Drive:https://drive.google.com/file/d/1YUhftGg9xip5Ff7qxtSo30fdnANXos1j/view?usp=sharing
- Baidu Yun:https://pan.baidu.com/s/1pzxIGRQI9vA4DZH8aXzZAw?pwd=6a1d
引用格式
bibtex @article{zhang2026acdis, title={Image Stitching in Adverse Condition: A Bidirectional-Consistency Learning Framework and Benchmark}, author={Zhang, Zengxi and Ge, JunChen and Jiang, Zhiying and Ma, Long and Liu, Jinyuan and Fan, Xin and Liu, Risheng}, journal={Advances in Neural Information Processing Systems}, year={2025}, }




