数据库所对应的标记软件是Vatic,官方网站地址为https://github.com/cvondrick/vatic
其中,groundtruth文件中对应的数据说明与vatic默认说明一致,中文解释如下:
txt文件中每一列对应物理意义:
1 Track ID. All rows with the same ID belong to the same path.
(视频标号,同一个id对应同一段视频)
2 xmin. The top left x-coordinate of the bounding box.
(目标矩形框的左上角x坐标)
3 ymin. The top left y-coordinate of the bounding box.
(目标矩形框的左上角y坐标)
4 xmax. The bottom right x-coordinate of the bounding box.
(目标矩形框的右下角x坐标)
5 ymax. The bottom right y-coordinate of the bounding box.
(目标矩形框的右下角y坐标)
6 frame. The frame that this annotation represents.
(相当于视频帧号,代表第几帧)
7 lost. If 1, the annotation is outside of the view screen.
(该项值为1说明目标没出现在当前帧)
8 occluded. If 1, the annotation is occluded.
(遮挡,该项在我们这次的标记中不存在,均为0,可以忽略)
9 generated. If 1, the annotation was automatically interpolated.
(该项为坐标及大小是软件自动线性计算而得的结果)
10 label. The label for this annotation, enclosed in quotation marks.
(目标类别标签,这里有三种,海参,海胆和扇贝)
11+ attributes. Each column after this is an attribute.
(类别标签之后紧跟着的属性,有appear和outside两种)
声明:
王璐瑶 大连理工大学信通学院 联系方式:luyaow@mail.dlut.edu.cn