CholecECA, RobustECA, PseudoECA
收藏arXiv2022-10-26 更新2024-06-21 收录
下载链接：
https://doi.org/10.7303/syn32148000, https://github.com/charliebudd/torch-content-area
下载链接
链接失效反馈官方服务：
资源简介：
本文介绍的数据集包括CholecECA、RobustECA和PseudoECA，这些数据集源自Cholec80和RobustMIS两个公开的腹腔镜视频数据集。数据集共包含6923个样本，其中3929个来自Cholec80，2994个来自RobustMIS。这些样本经过手动标注，用于提取内容区域边缘点候选。此外，PseudoECA数据集通过算法自动标注，用于训练学习边缘评分方法。这些数据集主要用于评估内容区域估计方法，特别是在腹腔镜手术视频处理和计算机视觉领域中的应用，旨在提高内容区域估计的准确性和实时性。
The datasets introduced herein include CholecECA, RobustECA, and PseudoECA, which are derived from two public laparoscopic video datasets: Cholec80 and RobustMIS. In total, these datasets contain 6,923 samples, among which 3,929 are sourced from Cholec80 and 2,994 from RobustMIS. All these samples are manually annotated for extracting candidate edge points of content regions. Additionally, the PseudoECA dataset is automatically annotated via algorithms, aiming to train edge scoring models. These datasets are primarily utilized to evaluate content region estimation methods, with applications particularly in laparoscopic surgical video processing and computer vision, with the goal of enhancing the accuracy and real-time performance of content region estimation.
提供机构：
伦敦国王学院
创建时间：
2022-10-26
