## Use Case and High-level Description

This is a MobileNetV2 + SSD-based vehicle and (Chinese) license plate detector for the "Barrier" use case.

## Specification

Metric Value
Mean Average Precision (mAP) 99.52%
AP vehicles 99.90%
AP plates 99.13%
Car pose Front facing cars
Min plate width 96 pixels
Max objects to detect 200
GFlops 0.271
MParams 0.547
Source framework TensorFlow*

Average Precision (AP) is defined as an area under the precision/recall curve. Validation dataset is BIT-Vehicle.

## Input

### Original Model

An input image, name: input, shape: 1, 256, 256, 3, format: B, H, W, C, where:

• B - batch size
• H - image height
• W - image width
• C - number of channels

Expected color order: RGB. Mean values: [127.5, 127.5, 127.5], scale factor for each channel: 127.5

### Converted Model

An input image, name: input, shape: 1, 3, 256, 256, format B, C, H, W, where:

• B - batch size
• C - number of channels
• H - image height
• W - image width

Expected color order is BGR.

## Output

### Original Model

The net outputs a blob with the shape: 1, 1, 200, 7 in the format 1, 1, N, 7, where N is the number of detected bounding boxes. For each detection, the description has the format: [image_id, label, conf, x_min, y_min, x_max, y_max], where:

• image_id - ID of the image in the batch
• label - predicted class ID
• conf - confidence for the predicted class
• (x_min, y_min) - coordinates of the top left bounding box corner
• (x_max, y_max) - coordinates of the bottom right bounding box corner.

### Converted Model

The net outputs a blob with the shape: 1, 1, 200, 7 in the format 1, 1, N, 7, where N is the number of detected bounding boxes. For each detection, the description has the format: [image_id, label, conf, x_min, y_min, x_max, y_max], where:

• image_id - ID of the image in the batch
• label - predicted class ID
• conf - confidence for the predicted class
• (x_min, y_min) - coordinates of the top left bounding box corner
• (x_max, y_max) - coordinates of the bottom right bounding box corner.

You can download models and if necessary convert them into Inference Engine format using the Model Downloader and other automation tools as shown in the examples below.