Hello Reshape SSD Python Sample¶
This sample demonstrates how to do synchronous inference of object detection models using the Shape Inference feature.
Models with only 1 input and output are supported.
The following Python API is used in the application:
Feature |
API |
Description |
---|---|---|
Model Operations |
openvino.runtime.Model.reshape , openvino.runtime.Model.input , openvino.runtime.Output.get_any_name , openvino.runtime.PartialShape |
Managing of model |
Basic OpenVINO™ Runtime API is described in Hello Classification Python Sample.
Options |
Values |
---|---|
Validated Models |
|
Validated Layout |
NCHW |
Model Format |
OpenVINO™ toolkit Intermediate Representation (.xml + .bin), ONNX (.onnx) |
Supported devices |
|
Other language realization |
How It Works¶
At startup, the sample application reads command-line parameters, prepares input data, loads a specified model and image to OpenVINO Runtime plugin, performs synchronous inference, and processes output data.
As a result, the program creates an output image, logging each step in a standard output stream.
For more information, refer to the explicit description of Integration Steps in the Integrate OpenVINO Runtime with Your Application guide.
Running¶
Before running the sample, specify a model and an image:
Use public or Intel’s pre-trained models from Open Model Zoo. The models can be downloaded by using the Model Downloader.
You may use images from the media files collection, available online in the test data storage.
To run the sample, use the following script:
python hello_reshape_ssd.py <path_to_model> <path_to_image> <device_name>
NOTES :
By default, samples and demos in OpenVINO Toolkit expect input with
BGR
order of channels. If you trained your model to work withRGB
order, you need to manually rearrange the default order of channels in the sample or demo application, or reconvert your model, using Model Optimizer with--reverse_input_channels
argument specified. For more information about the argument, refer to the When to Reverse Input Channels section of Embedding Preprocessing Computation.Before running the sample with a trained model, make sure that the model is converted to the OpenVINO Intermediate Representation format (*.xml + *.bin) by using Model Optimizer.
The sample accepts models in the ONNX format (.onnx) that do not require preprocessing.
Example¶
Install the
openvino-dev
Python package to use Open Model Zoo Tools:python -m pip install openvino-dev[caffe,onnx,tensorflow2,pytorch,mxnet]
Download a pre-trained model:
omz_downloader --name mobilenet-ssd
If a model is not in the OpenVINO IR or ONNX format, it must be converted. You can do this using the model converter:
omz_converter --name mobilenet-ssd
Perform inference of the
banana.jpg
, using thessdlite_mobilenet_v2
model on aGPU
, for example:python hello_reshape_ssd.py mobilenet-ssd.xml banana.jpg GPU
Sample Output¶
The sample application logs each step in a standard output stream and creates an output image, drawing bounding boxes for inference results with an over 50% confidence.
[ INFO ] Creating OpenVINO Runtime Core
[ INFO ] Reading the model: C:/test_data/models/mobilenet-ssd.xml
[ INFO ] Reshaping the model to the height and width of the input image
[ INFO ] Loading the model to the plugin
[ INFO ] Starting inference in synchronous mode
[ INFO ] Found: class_id = 52, confidence = 0.98, coords = (21, 98), (276, 210)
[ INFO ] Image out.bmp was created!
[ INFO ] This sample is an API example, for any performance measurements please use the dedicated benchmark_app tool