diff --git a/deploy/cpp/scripts/jetson_bootstrap.sh b/deploy/cpp/scripts/jetson_bootstrap.sh
new file mode 100644
index 0000000000000000000000000000000000000000..94d545a38abf1925567f54bc01afbba37371a5f3
--- /dev/null
+++ b/deploy/cpp/scripts/jetson_bootstrap.sh
@@ -0,0 +1,10 @@
+# download pre-compiled opencv lib
+OPENCV_URL=https://bj.bcebos.com/paddlex/deploy/tools/opencv3_aarch.tgz
+if [ ! -d "./deps/opencv3" ]; then
+    mkdir -p deps
+    cd deps
+    wget -c ${OPENCV_URL}
+    tar xvfj opencv3_aarch.tgz
+    rm -rf opencv3_aarch.tgz
+    cd ..
+fi
diff --git a/deploy/cpp/scripts/jetson_build.sh b/deploy/cpp/scripts/jetson_build.sh
new file mode 100644
index 0000000000000000000000000000000000000000..95bec3cac95be5cf686d63ec5b0f49f62e706586
--- /dev/null
+++ b/deploy/cpp/scripts/jetson_build.sh
@@ -0,0 +1,42 @@
+# 是否使用GPU(即是否使用 CUDA)
+WITH_GPU=OFF
+# 使用MKL or openblas
+WITH_MKL=OFF
+# 是否集成 TensorRT(仅WITH_GPU=ON 有效)
+WITH_TENSORRT=OFF
+# TensorRT 的路径，如果需要集成TensorRT，需修改为您实际安装的TensorRT路径
+TENSORRT_DIR=/root/projects/TensorRT/
+# Paddle 预测库路径, 请修改为您实际安装的预测库路径
+PADDLE_DIR=/root/projects/fluid_inference
+# Paddle 的预测库是否使用静态库来编译
+# 使用TensorRT时，Paddle的预测库通常为动态库
+WITH_STATIC_LIB=OFF
+# CUDA 的 lib 路径
+CUDA_LIB=/usr/local/cuda/lib64
+# CUDNN 的 lib 路径
+CUDNN_LIB=/usr/local/cuda/lib64
+
+# 是否加载加密后的模型
+WITH_ENCRYPTION=OFF
+
+# OPENCV 路径, 如果使用自带预编译版本可不修改
+sh $(pwd)/scripts/jetson_bootstrap.sh  # 下载预编译版本的opencv
+OPENCV_DIR=$(pwd)/deps/opencv3
+
+# 以下无需改动
+rm -rf build
+mkdir -p build
+cd build
+cmake .. \
+    -DWITH_GPU=${WITH_GPU} \
+    -DWITH_MKL=${WITH_MKL} \
+    -DWITH_TENSORRT=${WITH_TENSORRT} \
+    -DWITH_ENCRYPTION=${WITH_ENCRYPTION} \
+    -DTENSORRT_DIR=${TENSORRT_DIR} \
+    -DPADDLE_DIR=${PADDLE_DIR} \
+    -DWITH_STATIC_LIB=${WITH_STATIC_LIB} \
+    -DCUDA_LIB=${CUDA_LIB} \
+    -DCUDNN_LIB=${CUDNN_LIB} \
+    -DENCRYPTION_DIR=${ENCRYPTION_DIR} \
+    -DOPENCV_DIR=${OPENCV_DIR}
+make