Jetson nano DeepStream-5.1 YOLOv4

Download pytorch-YOLOv4

git clone https://github.com/Tianxiaomo/pytorch-YOLOv4.git
cd pytorch-YOLOv4

Download yolov4.cfg & yolov4.weight

wget https://raw.githubusercontent.com/AlexeyAB/darknet/master/cfg/yolov4.cfg
wget https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v3_optimal/yolov4.weights

Edit yolov4.cfg

[net]
#batch=64
#subdivisions=8
batch=1
subdivisions=1
# Training
#width=512
#height=512
#width=608
#height=608
width=416
height=416

Install torch

pip3 install torch
pip3 install torchvision
pip3 install onnxruntime

Install onnx

sudo apt-get install protobuf-compiler libprotoc-dev
pip3 install onnx -i https://pypi.doubanio.com/simple/

Enable swap

sudo fallocate -l 4.0G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile

Transform darknet to onnx

export OMP_NUM_THREADS=1
python3 demo_darknet2onnx.py yolov4.cfg yolov4.weights ./data/giraffe.jpg 1

Transform onnx to TensorRT engine

/usr/src/tensorrt/bin/trtexec --onnx=yolov4_1_3_416_416_static.onnx \
--explicitBatch --saveEngine=yolov4_1_3_416_416_fp16.engine \
--workspace=2048 --fp16

Download yolov4_deepstream

cd /opt/nvidia/deepstream/deepstream/sources/
git clone https://github.com/NVIDIA-AI-IOT/yolov4_deepstream.git
cd yolov4_deepstream/deepstream_yolov4

Copy TensorRT engine

cp $HOME/pytorch-YOLOv4/yolov4_1_3_416_416_fp16.engine .

Edit deepstream_app_config_yoloV4.txt

model-engine-file=yolov4_1_3_416_416_fp16.engine

Edit config_infer_primary_yoloV4.txt

model-engine-file=yolov4_1_3_416_416_fp16.engine
network-mode=2

Build

export CUDA_VER=10.0
make -C nvdsinfer_custom_impl_Yolo

Run

unset DISPLAY
rm -rf $HOME/.cache/gstreamer-1.0/registry.aarch64.bin

sudo route add -net 224.0.0.0 netmask 255.0.0.0 wlan9

deepstream-app -c deepstream_app_config_yoloV4.txt

Console output

Unknown or legacy key specified 'is-classifier' for group [property]
0:00:09.526815556 25328      0x21dc8d0 INFO                 nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::deserializeEngineAndBackend() <nvdsinfer_context_impl.cpp:1702> [UID = 1]: deserialized trt engine from :/opt/nvidia/deepstream/deepstream-5.1/sources/deepstream_yolov4/yolov4_1_3_416_416_fp16.engine
INFO: [Implicit Engine Info]: layers num: 3
0   INPUT  kFLOAT input           3x416x416       
1   OUTPUT kFLOAT boxes           10647x1x4       
2   OUTPUT kFLOAT confs           10647x80        
0:00:09.527013842 25328      0x21dc8d0 INFO                 nvinfer gstnvinfer.cpp:619:gst_nvinfer_logger:<primary_gie> NvDsInferContext[UID 1]: Info from NvDsInferContextImpl::generateBackendContext() <nvdsinfer_context_impl.cpp:1806> [UID = 1]: Use deserialized engine model: /opt/nvidia/deepstream/deepstream-5.1/sources/deepstream_yolov4/yolov4_1_3_416_416_fp16.engine
0:00:09.649444159 25328      0x21dc8d0 INFO                 nvinfer gstnvinfer_impl.cpp:313:notifyLoadModelStatus:<primary_gie> [UID 1]: Load new model:/opt/nvidia/deepstream/deepstream-5.1/sources/deepstream_yolov4/config_infer_primary_yoloV4.txt sucessfully
Runtime commands:
    h: Print this help
    q: Quit
    p: Pause
    r: Resume
**PERF:  FPS 0 (Avg)    
**PERF:  0.00 (0.00)    
** INFO: <bus_callback:181>: Pipeline ready
Opening in BLOCKING MODE
Opening in BLOCKING MODE 
NvMMLiteOpen : Block : BlockType = 261 
NVMEDIA: Reading vendor.tegra.display-size : status: 6 
NvMMLiteBlockCreate : Block : BlockType = 261 
** INFO: <bus_callback:167>: Pipeline running
**PERF:  4.92 (4.82)    
**PERF:  4.92 (4.92)    
**PERF:  4.92 (4.88)