LUNA SDK mobile models and dependencies

Feature name and
description

Required algorithms and used structures

Best Shot
Selection

Face Features
Extraction

Face
Recognition

The <device> placeholder shows that the model can be used with different devices (e. g., CPU, GPU).
Name of the model includes corresponding postfix (e. g., *_cpu.plan, *_gpu.plan).
See "Hardware requirements" section in "FaceEngine Handbook" for information about devices supported by NN.

Example of table using - if user wants to find best shots in his application, he needs to look at the "Best Shot Selection"
column and see what network models are required in order to run such functionality.
The required models and structures are marked with green.
Algorigthms and structures marked with red are not used for the functionality.
Face detection NN version 2
First, second and third NN
should be launched consequently		 Structure(s):
IDetector

Plan(s):
FaceDet_v2_first_<device>.plan
FaceDet_v2_second_<device>.plan
FaceDet_v2_third_<device>.plan
BestShotQuality estimation
Performs head pose and AGS estimations		 Structure(s):
IBestShotQualityEstimator

Plan(s):
ags_v3_<device>.plan
headpose_v3_<device>.plan
Image quality estimation
(light, dark, blur, etc.)		 Structure(s):
IQualityEstimator

Plan(s):
model_subjective_quality_<version>_<device>.plan
Image glasses estimation
(light, dark, blur, etc.)		 Structure(s):
IGlassesEstimator

Plan(s):
glasses_estimation_v2_<version>_<device>.plan
Head pose estimation Structure(s):
IHeadPoseEstimator

Plan(s):
headpose_v3_<device>.plan
AGS estimation Structure(s):
IAGSEstimator

Plan(s):
ags_v3_<device>.plan
LivenessOneShotRGB estimation
Predicts whether the person's face is real or fake (photo, printed image)		 Structure(s):
ILivenessOneShotRGBEstimator

Plan(s):
oneshot_rgb_liveness_<version>_model_<number>_<device>.plan

v12 plans are bundled by default, while v11 plans are bundled in a separate archive
NIRLiveness estimation
Predicts whether the person's face is real or fake (photo, printed image)		 Structure(s):
INIRLivenessEstimator

Plan(s):
nir_liveness_<version>_model_<number>_<device>.plan
Extraction (NN version 60 mobilenet)
(Available in Complete distribution package only!) 		Structure(s):
IDescriptor, IDescriptorBatch, IDescriptorExtractor

Plan(s):
cnn60m_<device>.plan
Eyes state estimation (eyes status and
keypoints for irises and eyelids)		 Structure(s):
IEyeEstimator

Plan(s):
eyes_estimation_flwr8_<device>.plan
Eyes state
(lightened version, returns eyes state only)		 Structure(s):
IEyeEstimator

Plan(s):
eye_status_estimation_<device>.
Luna SDK functionality

Functionality

Feature name and
description

Required algorithms and used structures

Face detection NN
versions 1 and 2

Face detection NN
version 3
MedicalMaskDetection Medical mask estimation Structure(s):
IMedicalMaskEstimator

Plan(s):
mask_clf_<version>_<device>.plan
Mouth Estimation Mouth estimation Structure(s):
IMouthEstimatorEstimator

Plan(s):
mouth_estimation_<version>_<device>.plan
FaceOcclusion Estimation FaceOcclusion estimation Structure(s):
IFaceOcclusionEstimator

Plan(s):
face_occlusion_<version>_<device>.plan
TrackEngine TrackEngine Structure(s):
ITrackEngine

Plan(s):
vlTracker_detection_<device>.plan
vlTracker_template_<device>.plan
vlTracker_update_<device>.plan

The table above includes separately licensed feature.

Additional features required for the functionality utilization are shown in the columns to the right.