I measured the inference speed of each TensorFlow Lite Detection Model on a Raspberry Pi by following the instructions from the following link:
https://www.ejtech.io/learn/tflite-object-detection-model-comparison. On the website there is a series of commands to download and run the benchmark tool on the Pi:
Commands to download and run TFLite benchmark tool
cd ~/tflite1
wget -O benchmark_model https://storage.googleapis.com/tensorflow-nightly-public/prod/tensorflow/release/lite/tools/nightly/latest/linux_arm_benchmark_model_plus_flex
sudo chmod +x benchmark_model
./benchmark_model --graph=<graph_name>/detect.tflite --num_threads=4 --warmup_runs=10 --num_runs=100
However, because my raspberry pi uses aarch64, I changed it to:
wget -O benchmark_model https://storage.googleapis.com/tensorflow-nightly-public/prod/tensorflow/release/lite/tools/nightly/latest/linux_aarch64_benchmark_model_plus_flex
I executed it on rasp terminal:
$ sudo chmod +x benchmark_model
$ ./benchmark_model --graph=Sample_TFLite_model/detect.tflite --num_threads=4 --warmup_runs=10 --num_runs=100
The Result:
./benchmark_model: /lib/aarch64-linux-gnu/libstdc++.so.6: version CXXABI_1.3.13 not found (required by ./benchmark_model)
./benchmark_model: /lib/aarch64-linux-gnu/libstdc++.so.6: version GLIBCXX_3.4.29 not found (required by ./benchmark_model)
./benchmark_model: /lib/aarch64-linux-gnu/libc.so.6: version GLIBC_2.34 not found (required by ./benchmark_model)
./benchmark_model: /lib/aarch64-linux-gnu/libc.so.6: version GLIBC_2.32 not found (required by
./benchmark_model) ./benchmark_model: /lib/aarch64-linux-gnu/libc.so.6: version GLIBC_2.33 not found (required by ./benchmark_model)