Neural networks used in LUNA ID#
In LUNA ID, neural networks efficiently and accurately process faces in both images and video streams. These neural networks are stored in .plan files.
The table below lists all .plan files used in LUNA ID, along with the functionalities they provide. Some of these files are required for integrating LUNA ID into your application.
Note, that using the .plan files will add extra size to your app. To learn how to exclude extra .plan files, see Reducing your app size by excluding .plan files.
| .plan file | OS | Size, MB | Required | Functionality |
|---|---|---|---|---|
| ags_v3_arm.plan | 0,62 | Best shot quality estimation | ||
| cnn60m_arm.plan | 18,54 | Descriptor generation from an image See also: |
||
| eye_status_estimation_arm.plan | 0,26 | Eye state estimation See also: |
||
| eyes_estimation_flwr8_arm.plan | 0,94 | Eye state estimation See also: |
||
| face_occlusion_v1_arm.plan | 0,17 | Face occlusion | ||
| FaceDet_v2_first_arm.plan FaceDet_v2_second_arm.plan FaceDet_v2_third_arm.plan |
|
0,01 0,11 1,64 |
|
Face detection See also: |
| glasses_estimation_v2_arm.plan | 0,72 | Glasses estimation See also: |
||
| headpose_v3_arm.plan | 0,28 | Head pose estimation | ||
| mask_clf_v3_arm-int8.plan | 2,64 | Medical mask estimation See also: |
||
| model_subjective_quality_v1_arm.plan model_subjective_quality_v2_arm.plan |
|
0,05 0,38 |
|
Image quality estimation See also: |
| mouth_estimator_v4_arm.plan | 1,56 | Mouth estimation | ||
| oneshot_rgb_liveness_v12_model_4_arm.plan oneshot_rgb_liveness_v12_model_5_arm.plan oneshot_rgb_liveness_v12_model_6_arm.plan |
|
4 4 4,64 |
Offline OneShotLiveness estimation See also: |