Sphereface

Use Case and High-Level Description

Deep face recognition under open-set protocol

Specification

Metric

Value

Type

Face recognition

GFLOPs

3.504

MParams

22.671

Source framework

Caffe*

Accuracy

Metric

Value

LFW accuracy

98.8321%

Input

Original model

Image, name - data, shape - 1, 3, 112, 96, format is B, C, H, W, where:

  • B - batch size

  • C - channel

  • H - height

  • W - width

Channel order is BGR. Mean values - [127.5, 127.5, 127.5], scale value - 128

Converted model

Image, name - data, shape - 1, 3, 112, 96, format is B, C, H, W, where:

  • B - batch size

  • C - channel

  • H - height

  • W - width

Channel order is BGR.

Output

Original model

Face embeddings, name - fc5, shape - 1, 512, output data format - B, C, where:

  • B - batch size

  • C - row-vector of 512 floating points values, face embeddings

The net outputs on different images are comparable in cosine distance.

Converted model

Face embeddings, name - fc5, shape - 1, 512, output data format - B, C, where:

  • B - batch size

  • C - row-vector of 512 floating points values, face embeddings

The net outputs on different images are comparable in cosine distance.

Download a Model and Convert it into Inference Engine Format

You can download models and if necessary convert them into Inference Engine format using the Model Downloader and other automation tools as shown in the examples below.

An example of using the Model Downloader:

python3 <omz_dir>/tools/downloader/downloader.py --name <model_name>

An example of using the Model Converter:

python3 <omz_dir>/tools/downloader/converter.py --name <model_name>