Closed guillebot closed 4 years ago
If you are on the latest JetPack(4.3) ,use this command sudo pip3 install --extra-index-url https://developer.download.nvidia.com/compute/redist/jp/v43 tensorflow==1.15.2+nv20.2
to install tensorflow and check https://docs.nvidia.com/deeplearning/frameworks/install-tf-jetson-platform/index.html to see if you have all required packages for tf and then you should be good to go ( actually tested the FP32 model and i'm running at 50 FPS on my Xavier)
Thank you very much for your time.
Np :) Running at 100 FPS with INT8 Precision now, amazing
May I ask the camera specs? resolution, colors, bit depth?
I intend to run it on a couple of e-con systems 4k cameras.
thanks
On Mon, Mar 9, 2020 at 10:32 AM Kmarconi notifications@github.com wrote:
Np :) Running at 100 FPS with INT8 Precision now, amazing
— You are receiving this because you modified the open/close state. Reply to this email directly, view it on GitHub https://github.com/NVIDIA/object-detection-tensorrt-example/issues/15?email_source=notifications&email_token=AAIXVWMUXZXPA6KT47TL2DTRGTVW3A5CNFSM4KPZ5RVKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOEOHEFFA#issuecomment-596525716, or unsubscribe https://github.com/notifications/unsubscribe-auth/AAIXVWIGH2G2P6XQW75DLCTRGTVW3ANCNFSM4KPZ5RVA .
Only one 1080p camera for me
I know it's a different architecture but is it possible to install tensorrt+opencv.
Any hints about how to port this sample?
thanks