CVAT integration feedback

[mkutu]

some notes on CVAT experience thus far before i forget

• allow users to add images to a task that has already been created. It seems now you can only add images when you create the task.
• prohibit users from leaving "create task" page before images have moved to server. I did this once and the images never showed up in the task option screen.
• provide multiple export options at main task page after clicking (Export task dataset). As of now, this is only available if you choose the "job" and enter the annotation page. It would be easier to not have to wait for the page to load. I guess the annotations would have to be reprocessed on the backend to conform to a new export format.

This is a really great tool. And I'm especially liking the lack of lag, and opencv tool. Keep up the good work and thanks for including us in the testing.