• Hardware Platform (Jetson / GPU) Jetson Nano
• DeepStream Version 6.0.1
• JetPack Version (valid for Jetson only) 4.6.1
When running the “deepstream-emotion-app” from NVIDIA-AI-IOT/deepstream_tao_apps with the command: “/dli/task/deepstream_tao_apps/apps/tao_others/deepstream-emotion-app# ./deepstream-emotion-app 2 …/…/…/configs/facial_tao/sample_faciallandmarks_config.txt file:///dli/task/deepstream_tao_apps/apps/tao_others/deepstream-emotion-app/face_test.jpg ./landmarks” i get the following output in the end: “[MemUsageChange] Init CUDA: CPU +0, GPU +0, now: CPU 918, GPU 3467 (MiB)
Loaded engine size: 4 MiB
Using cublas as a tactic source
[MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +0, now: CPU 920, GPU 3469 (MiB)
Using cuDNN as a tactic source
[MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 920, GPU 3469 (MiB)
Deserialization required 65653 microseconds.
[MemUsageChange] TensorRT-managed allocation in engine deserialization: CPU +0, GPU +4, now: CPU 3, GPU 202 (MiB)
Using cublas as a tactic source
[MemUsageChange] Init cuBLAS/cuBLASLt: CPU +0, GPU +0, now: CPU 920, GPU 3469 (MiB)
Using cuDNN as a tactic source
[MemUsageChange] Init cuDNN: CPU +0, GPU +0, now: CPU 920, GPU 3469 (MiB)
Total per-runner device persistent memory is 0
Total per-runner host persistent memory is 7296
Allocated activation device memory of size 217088
[MemUsageChange] TensorRT-managed allocation in IExecutionContext creation: CPU +0, GPU +1, now: CPU 3, GPU 203 (MiB)
Average fps 0.000233
Totally 0 faces are inferred
Deleting pipeline”
The image i’m using its litereally a guy’s face so i don’t understand the “Totally 0 faces are inferred” result. Beside that the average fps is extremely low. Any way to be able to use this pipeline properly with Jetson Nano? Maybe its to heavy for it’s hardware?