MobileNet SSD Object Detection Test

generate-mobilenet-ssd-model.sh

outdir=mobilenet-ssd

# Download caffe models into tmp
sdk/deployment_tools/model_downloader/downloader.py --name mobilenet-ssd -o $(outdir)

# To know the parameters for the model optimizer, look for "model_optimizer_args" in "mobilenet-ssd" section in "list_topologies.yml". It is in the same directory as downloader.py.
sdk/deployment_tools/model_optimizer/mo.py --framework caffe --data_type FP32 --input_shape [1,3,300,300] --input data --mean_values data[127.5,127.5,127.5] --scale_values data[127.50223128904757] --output detection_out --input_model tmp/object_detection/common/mobilenet-ssd/caffe/mobilenet-ssd.caffemodel --input_proto tmp/object_detection/common/mobilenet-ssd/caffe/mobilenet-ssd.prototxt -o $(outdir)/FP32
sdk/deployment_tools/model_optimizer/mo.py --framework caffe --data_type FP16 --input_shape [1,3,300,300] --input data --mean_values data[127.5,127.5,127.5] --scale_values data[127.50223128904757] --output detection_out --input_model tmp/object_detection/common/mobilenet-ssd/caffe/mobilenet-ssd.caffemodel --input_proto tmp/object_detection/common/mobilenet-ssd/caffe/mobilenet-ssd.prototxt -o $(outdir)/FP16

mobilenet-ssd-cpu.sh

# specify "-i cam" to read from the first camera, or "-i <file or URL>" to read from a stream.
bin/object_detection_demo_ssd_async -m mobilenet-ssd/FP32/mobilenet-ssd.xml -d CPU $*

mobilenet-ssd-myriad.sh

# specify "-i cam" to read from the first camera, or "-i <file or URL>" to read from a stream.
bin/object_detection_demo_ssd_async -m mobilenet-ssd/FP16/mobilenet-ssd.xml -d MYRIAD $*

mobilenet-ssd.labels

background
aeroplane
bicycle
bird
boat
bottle
bus
car
cat
chair
cow
diningtable
dog
horse
motorbike
person
pottedplant
sheep
sofa
train
tvmonitor