person-detection-retail-0002

Use Case and High-Level Description

This is a pedestrian detector based on backbone with hyper-feature + R-FCN for the Retail scenario.

Example

Specification

Metric Value
AP 80.14%
Pose coverage Standing upright, parallel to image plane
Support of occluded pedestrians YES
Occlusion coverage <50%
Min pedestrian height 80 pixels (on 1080p)
Max objects to detect 200
GFlops 12.427
MParams 3.244
Source framework Caffe*

Average Precision (AP) is defined as an area under the precision/recall curve. Validation dataset consists of ~50K of images from ~100 different scenes.

Inputs

  1. Image, name: data, shape: 1, 3, 544, 992 in format 1, C, H, W, where:

    • C - number of channels
    • H - image height
    • W - image width

    The expected channel order is BGR.

  2. name: im_info, shape: 1x6 - An image information [544, 992, 992/frame_width, 544/frame_height, 992/frame_width, 544/frame_height]

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 (1 - 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

Legal Information

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