Сaption near bounding box

As of now, detectnet_v2 inference does not support drawing caption directly on the inferenced images. You can run “tao detectnet_v2 inference” to get the inferenced images/labels and then leverage the label to draw caption.