landmarks-regression-retail-0009

Use Case and High-Level Description

This is a lightweight landmarks regressor for the Smart Classroom scenario. It has a classic convolutional design: stacked 3x3 convolutions, batch normalizations, PReLU activations, and poolings. Final regression is done by the global depthwise pooling head and FullyConnected layers. The model predicts five facial landmarks: two eyes, nose, and two lip corners.

Example

_images/landmarks-regression-retail-0009.png

Specification

Metric

Value

Mean Normed Error (on VGGFace2)

0.0705

Face location requirements

Tight crop

GFlops

0.021

MParams

0.191

Source framework

PyTorch*

Normed Error (NE) for i th sample has the following form:

_images/normed_error.png

where N is the number of landmarks, p -hat and p are, correspondingly, the prediction and ground truth vectors of k th landmark of i th sample, and d i is the interocular distance for i th sample.

Inputs

Image, name: 0, shape: 1, 3, 48, 48 in the format B, C, H, W, where:

  • B - batch size

  • C - number of channels

  • H - image height

  • W - image width

The expected color order is BGR.

Outputs

The net outputs a blob with the shape: 1, 10, 1, 1, containing a row-vector of 10 floating point values for five landmarks coordinates in the form (x0, y0, x1, y1, …, x4, y4). All the coordinates are normalized to be in range [0, 1].

Demo usage

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