Summary
Difference between models when running OpenVINO™ toolkit sample on Edge Insights for Autonomous Mobile Robots (AMR)
Description
OpenVINO™ toolkit sample results in different frames per second (FPS) when inferencing model after using the Model Optimizer to convert to Intermediate Representation (IR) format.
Resolution
The OpenVINO™ toolkit sample is inferencing two different models and the performance difference is expected.
- vehicle-license-plate-detection-barrier-0106 is downloaded and converted from the Open Model Zoo
- ssd_resnet50_v1_fpn is downloaded from Tensorflow and is converted using the Model Optimizer