person-detection-0202

Use Case and High-Level Description

This is a person detector that is based on MobileNetV2 backbone with two SSD heads from 1/16 and 1/8 scale feature maps and clustered prior boxes for 512x512 resolution.

Example

Specification

Metric

Value

AP @ [ IoU=0.50:0.95 ]

0.322 (internal test set)

GFlops

3.143

MParams

1.817

Source framework

PyTorch*

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

Inputs

Image, name: image, shape: 1, 3, 512, 512 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.

Outputs

The net outputs blob with shape: 1, 1, 200, 7 in the format 1, 1, N, 7, where N is the number of detected bounding boxes. Each detection 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 (0 - person)

  • 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.

Training Pipeline

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

Demo usage

The model can be used in the following demos provided by the Open Model Zoo to show its capabilities: