face-detection-0206¶
Use Case and High-Level Description¶
Face detector based on ResNet152 as a backbone with a ATSS head for indoor and outdoor scenes shot by a front-facing camera.
Example¶
Specification¶
Metric |
Value |
---|---|
AP ( WIDER ) |
94.27% |
GFlops |
339.602 |
MParams |
69.920 |
Source framework |
PyTorch* |
Average Precision (AP) is defined as an area under the precision/recall curve. All numbers were evaluated by taking into account only faces bigger than 64 x 64 pixels.
Inputs¶
Image, name: image
, shape: 1, 3, 640, 640
in the format B, C, H, W
, where:
B
- batch sizeC
- number of channelsH
- image heightW
- image width
Expected color order: BGR
.
Outputs¶
The
boxes
is a blob with the shape750, 5
in the formatN, 5
, whereN
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 cornerconf
- confidence for the predicted class
The
labels
is a blob with the shape750
in the formatN
, whereN
is the number of detected bounding boxes. It contains predicted class ID (0 - face) per each detected box.
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:
Legal Information¶
[*] Other names and brands may be claimed as the property of others.