Camera Capture with TensorRT and Multi-Channel Encoding - Sample

Hello,

where can i find the Source Code for the “Camera Capture with TensorRT and Multi-Channel Encoding” Sample from the L4TMultimediaAPIReference?

Hi,

Please install MMAPI via Jetpack.

04_video_dec_gie and backend are the samples of tensorRT.

$ ./04_video_dec_gie/video_dec_gie ../data/video/sample_outdoor_car_1080p_10fps.h264 H264 \
                --gie-deployfile ../data/model/GoogleNet-modified.prototxt \
                --gie-modelfile ../data/model/GoogleNet-modified-online_iter_30000.caffemodel \
                --gie-forcefp32 0 --gie-enable-perf 1
$ ./backend/backend 1 ../data/video/sample_outdoor_car_1080p_10fps.h264 H264 \
                --gie-deployfile ../data/model/GoogleNet-modified.prototxt \
                --gie-modelfile ../data/model/GoogleNet-modified-online_iter_30000.caffemodel \
                --gie-forcefp32 0 --gie-proc-interval 1 -fps 10

Hi,

thanks for your reply! Okay with these examples i can do tensorRT from Video Files.

But what i’m looking for is the example for Camera Capture + tensorRT + multichannel Video Encoding. In the L4TMultimediaAPIReference you can find the Example under “L4TMultimediaAPIReference/nvee_docs/ee_camcap_tensorrt_multichannel_group.html”. The Execution File is called “./frontend [OPTIONS]”. Where can i find this Source Code?

Hi,

Thanks for your question.
We provide frontend sample in rel-27.1 branch, which can be found in JetPack3.0.

Hi,

Thanks! So the frontend sample is only working with Jetson TX2, right?

Hi,

Yes, it’s only for tx2.
But the source code still can give you some hint.

Thanks.