Please provide the following information when requesting support.
• Hardware (T4/V100/Xavier/Nano/etc) RTX GPU
• Network Type (Detectnet_v2/Faster_rcnn/Yolo_v4/LPRnet/Mask_rcnn/Classification/etc) BodyPose2d
• TLT Version (Please run “tlt info --verbose” and share “docker_tag” here)
• Training spec file(If have, please share here)
• How to reproduce the issue ? (This is for errors. Please share the command line and the detailed log here.)
Hi. I would like to ask a question referring to the code below.
My purpose was to parse only the 2.5d output layer, map it from 2d to 3d using the intrinsic parameters of my Rtsp camera obtained in advance (camera calibration), and then proceed with the Pose classification task.
So, in a way, the output directly used is pose25d, but the input stage is In addition to the image, do the four parameters need to be accurate? I think pose25d is only used with heatmap information anyway.
5 posts - 3 participants