GooseDetect^lion

The GooseDetect^lion is the first large lion-head goose detection dataset, which consists of 2,660 images and 98,111 bounding box annotations. The dataset was collected with 6 cameras deployed in a goose farm in Chenghai district of Shantou city, Guangdong province. Images sampled from videos collected during July 9 -10 in 2022 were fully annotated by a team of fifty volunteers. The data were collected by 6 cameras deployed in a lion-head goose farm. The dataset contains 2 categories of geese defined chronologically: big geese and small geese. The big geese are older than 9 weeks, and the small ones are between 5 and 8 weeks. It is fully and manually annotated at the frame level, with following labels:• Bounding boxes: The position of each goose in an image is given in the form of a bounding box.• Goose category tags: Each goose is labelled by a goose category tag, i.e., “big goose” or “small goose”.Our dataset was uploaded in VOC2007 format, which contains 2660 images(.jpg) and corresponding annotation files(.xml).

搜集汇总

背景与挑战

背景概述

GooseDetect^lion是首个大型狮头鹅检测数据集，包含2,660张图像和98,111个边界框标注，采集自广东汕头鹅场的监控视频。数据按鹅的年龄分为大鹅(9周+)和小鹅(5-8周)两类，采用VOC2007格式存储。