Use Case and High-Level Description

This is a person, vehicle, bike detector that is based on MobileNetV2 backbone with ATSS head for 448x256 resolution.



Metric Value
AP @ [ IoU=0.50:0.95 ] 0.274 (internal test set)
GFlops 1.811
MParams 2.327
Source framework PyTorch*

Average Precision (AP) is defined as an area under the precision/recall curve.


Image, name: input, shape: 1, 3, 256, 448 in the format B, C, H, W, where:

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

Expected color order is BGR.


  1. The boxes is a blob with the shape 100, 5 in the format N, 5, where N is the number of detected bounding boxes. For each detection, the description has the format: [x_min, y_min, x_max, y_max, conf], where:
    • (x_min, y_min) - coordinates of the top left bounding box corner
    • (x_max, y_max) - coordinates of the bottom right bounding box corner
    • conf - confidence for the predicted class
  2. The labels is a blob with the shape 100 in the format N, where N is the number of detected bounding boxes. The value of each label is equal to predicted class ID (0 - vehicle, 1 - person, 2 - non-vehicle).

Training Pipeline

The OpenVINO Training Extensions provide a training pipeline, allowing to fine-tune the model on custom dataset.

Legal Information

[*] Other names and brands may be claimed as the property of others.