As i said above, i annotation one video in the web, and when i dump the video's result with 'txt' file type, objects in the result was annotationed twice even more.
For example, there are only 2 person in the image frame(p1:[10,20,30,40], p2:[10,30,50,70]), but actually in the txt file outputed, there are 8 notes even more。
Here is snapshot of the partial result，you can find some person was annotationed twice even more.
As i said above, i annotation one video in the web, and when i dump the video's result with 'txt' file type, objects in the result was annotationed twice even more. For example, there are only 2 person in the image frame(p1:[10,20,30,40], p2:[10,30,50,70]), but actually in the txt file outputed, there are 8 notes even more。
Here is snapshot of the partial result，you can find some person was annotationed twice even more.
Look forward to your answer! Thanks.