遇见数据集

CTG2.0

收藏
Zenodo2026-07-20 更新2026-08-01 收录
官方服务:

资源简介:

CTG 2.0 is a vessel-aware laboratory vision benchmark consisting of 3,668 annotated laboratory images with polygon-based YOLO segmentation labels split into training, validation and test subsets. The dataset covers 30 object categories in total: 22 types of lab glass containers (beaker, burette, conical flask, volumetric flask, separatory funnel, test tube and other common glassware), four multiphase interface classes (gas-liquid, gas-solid, liquid-liquid, liquid-solid interfaces), plus four auxiliary categories including corks, labels and solid-solid interfaces. It targets the challenging task of thin elongated phase boundary segmentation inside translucent lab vessels for chemical process monitoring

提供机构:
pselab
创建时间:
2026-07-20
二维码
社区交流群
二维码
科研交流群
商业服务