Enable Face and Face Attribute Recognition

FindFace Server allows you to recognize human faces and face attributes. Subject to your needs, you can enable recognition of such face attributes as age, gender, emotions, beard, glasses, medical masks, head position, or liveness.

Face and face attribute recognition can be manually enabled and configured. Face and face attribute recognition works on both GPU- and CPU-acceleration.

Note

There are examples of how the sections should look like below. It may vary depending on the models that you have selected.

To enable face recognition, do the following:

Specify neural network models for face object detection in the /opt/ffserver/configs/extraction-api.yaml configuration file.

Important

Be sure to choose the right acceleration type for each model, matching the acceleration type of extraction-api: CPU or GPU. Be aware that extraction-api on CPU can work only with CPU-models, while extraction-api on GPU supports both CPU- and GPU-models.

Open the extraction-api.yaml configuration file.

sudo vi /opt/ffserver/configs/extraction-api.yaml

Specify the face detector model in the detectors -> models section by pasting the following code:

GPU

detectors:
  ...
  models:
    ...
    face_jasmine:
      aliases:
      - face
      - nnd
      - cheetah
      model: detector/facedet.jasmine_fast.004.gpu.fnk
      options:
        min_object_size: 32
        resolutions:
        - 2048x2048
  ...

CPU

detectors:
  ...
  models:
    ...
    face_jasmine:
      aliases:
      - face
      - nnd
      - cheetah
      model: detector/facedet.jasmine_fast.004.cpu.fnk
      options:
        min_object_size: 32
        resolutions:
        - 2048x2048
  ...

Make sure that the objects -> face section contains the quality_attribute: face_quality and the base_normalizer: facenorm/crop2x.v2_maxsize400.gpu.fnk or the base_normalizer: facenorm/crop2x.v2_maxsize400.cpu.fnk, depending on your acceleration type:

GPU

objects:
  ...
  face:
    base_normalizer: facenorm/crop2x.v2_maxsize400.gpu.fnk
    quality_attribute: face_quality
  ...

CPU

objects:
  ...
  face:
    base_normalizer: facenorm/crop2x.v2_maxsize400.cpu.fnk
    quality_attribute: face_quality
  ...

Specify the face normalizer models in the normalizers section by pasting the following code:

GPU

normalizers:
  ...
  models:
    crop1x:
      model: facenorm/crop1x.v2_maxsize400.gpu.fnk
    crop2x:
      model: facenorm/crop2x.v2_maxsize400.gpu.fnk
    cropbbox:
      model: facenorm/cropbbox.v2.gpu.fnk
    multicrop_full_center:
      model: ''
    multicrop_full_crop2x:
      model: facenorm/facenorm.multicrop_full_crop2x_size400.gpu.fnk
    norm200:
      model: facenorm/bee.v3.gpu.fnk
  ...

CPU

normalizers:
  ...
  models:
    crop1x:
      model: facenorm/crop1x.v2_maxsize400.cpu.fnk
    crop2x:
      model: facenorm/crop2x.v2_maxsize400.cpu.fnk
    cropbbox:
      model: facenorm/cropbbox.v2.cpu.fnk
    multicrop_full_center:
      model: ''
    multicrop_full_crop2x:
      model: facenorm/facenorm.multicrop_full_crop2x_size400.cpu.fnk
    norm200:
      model: facenorm/bee.v3.cpu.fnk
  ...

Note

This step is required to enable face attribute recognition.

To enable face attribute recognition, do the following:

In the /opt/ffserver/configs/extraction-api.yaml configuration file, specify the extraction models in the extractors section, subject to the extractors you want to enable. Be sure to indicate the right acceleration type for each model, matching the acceleration type of extraction-api: CPU or GPU.

GPU

extractors:
  ...
  models:
    face_age:
      default:
        model: faceattr/faceattr.age.v3.gpu.fnk
    face_beard:
      default:
        model: faceattr/beard.v0.gpu.fnk
    face_emben:
      default:
        model: face/nectarine_xl_320.gpu.fnk
    face_emotions:
      default:
        model: faceattr/emotions.v1.gpu.fnk
    face_gender:
      default:
        model: faceattr/faceattr.gender.v3.gpu.fnk
    face_glasses3:
      default:
        model: faceattr/glasses3.v0.gpu.fnk
    face_headpose:
      default:
        model: faceattr/headpose.v3.gpu.fnk
    face_liveness:
      default:
        model: faceattr/faceattr.liveness_web.v1.gpu.fnk
    face_medmask3:
      default:
        model: faceattr/medmask3.v2.gpu.fnk
    face_quality:
      default:
         model: faceattr/faceattr.quality.v5.gpu.fnk
  ...

CPU

extractors:
  ...
  models:
    face_age:
      default:
        model: faceattr/faceattr.age.v3.cpu.fnk
    face_beard:
      default:
        model: faceattr/beard.v0.cpu.fnk
    face_emben:
      default:
        model: face/nectarine_xl_320.cpu.fnk
    face_emotions:
      default:
        model: faceattr/emotions.v1.cpu.fnk
    face_gender:
      default:
        model: faceattr/faceattr.gender.v3.cpu.fnk
    face_glasses3:
      default:
        model: faceattr/glasses3.v0.cpu.fnk
    face_headpose:
      default:
        model: faceattr/headpose.v3.cpu.fnk
    face_liveness:
      default:
        model: faceattr/faceattr.liveness_web.v1.cpu.fnk
    face_medmask3:
      default:
        model: faceattr/medmask3.v2.cpu.fnk
    face_quality:
      default:
        model: faceattr/faceattr.quality.v5.cpu.fnk
  ...

The most used extraction models are the following:

Extractor	Acceleration	Configure as follows
age	CPU	`face_age: faceattr/faceattr.age.v3.cpu.fnk`
age	GPU	`face_age: faceattr/faceattr.age.v3.gpu.fnk`
beard	CPU	`face_beard: faceattr/beard.v0.cpu.fnk`
	GPU	`face_beard: faceattr/beard.v0.gpu.fnk`
	CPU	`face_beard4: faceattr/faceattr.beard4.v1.cpu.fnk`
	GPU	`face_beard4: faceattr/faceattr.beard4.v1.cpu.fnk`
individual face feature vector	CPU	`face_emben: face/nectarine_xl_320.cpu.fnk`
individual face feature vector	GPU	`face_emben: face/nectarine_xl_320.gpu.fnk`
gender	CPU	`face_gender: faceattr/faceattr.gender.v3.cpu.fnk`
gender	GPU	`face_gender: faceattr/faceattr.gender.v3.gpu.fnk`
emotions	CPU	`face_emotions: faceattr/emotions.v1.cpu.fnk`
emotions	GPU	`face_emotions: faceattr/emotions.v1.gpu.fnk`
glasses	CPU	`face_glasses3: faceattr/glasses3.v0.cpu.fnk`
	GPU	`face_glasses3: faceattr/glasses3.v0.gpu.fnk`
	CPU	`face_glasses4: faceattr/faceattr.glasses4.v0.cpu.fnk`
head position	CPU	`face_headpose: faceattr/headpose.v3.cpu.fnk`
head position	GPU	`face_headpose: faceattr/headpose.v3.gpu.fnk`
face liveness	CPU	`face_liveness: faceattr/faceattr.liveness_web.v1.cpu.fnk`
face liveness	GPU	`face_liveness: faceattr/faceattr.liveness_web.v1.gpu.fnk`
face mask	CPU	`face_medmask3: faceattr/medmask3.v2.cpu.fnk`
	GPU	`face_medmask3: faceattr/medmask3.v2.gpu.fnk`
	CPU	`face_medmask4: faceattr/faceattr.medmask4.v0.cpu.fnk`
	GPU	`face_medmask4: faceattr/faceattr.medmask4.v0.gpu.fnk`
face quality	CPU	`face_quality: faceattr/faceattr.quality.v5.cpu.fnk`
face quality	GPU	`face_quality: faceattr/faceattr.quality.v5.gpu.fnk`

To enable face recognition, modify the /opt/ffserver/configs/video-worker.yaml configuration file.

In the models section, specify the face neural network models by analogy with the example below:

GPU

sudo vi /opt/ffserver/configs/video-worker.yaml

models:
  ...
  detectors:
    ...
    face:
      fnk_path: /usr/share/findface-data/models/detector/facedet.jasmine_fast.004.gpu.fnk
      min_size: 60
    ...
  normalizers:
    ...
    face_norm:
      fnk_path: /usr/share/findface-data/models/facenorm/crop2x.v2_maxsize400.gpu.fnk
    face_norm_quality:
      fnk_path: /usr/share/findface-data/models/facenorm/crop1x.v2_maxsize400.gpu.fnk
    ...
  extractors:
    ...
    face_quality:
      fnk_path: /usr/share/findface-data/models/faceattr/faceattr.quality.v5.gpu.fnk
      normalizer: face_norm_quality
    ...

CPU

models:
  ...
  detectors:
    ...
    face:
      fnk_path: /usr/share/findface-data/models/detector/facedet.jasmine_fast.004.cpu.fnk
      min_size: 60
    ...
  normalizers:
    ...
    face_norm:
      fnk_path: /usr/share/findface-data/models/facenorm/crop2x.v2_maxsize400.cpu.fnk
    face_norm_quality:
      fnk_path: /usr/share/findface-data/models/facenorm/crop1x.v2_maxsize400.cpu.fnk
    ...
  extractors:
    ...
    face_quality:
      fnk_path: /usr/share/findface-data/models/faceattr/faceattr.quality.v5.cpu.fnk
      normalizer: face_norm_quality
    ...

Make sure that the objects -> face section is included:

objects:
  ...
  face:
    normalizer: face_norm
    quality: face_quality
    track_features: ''

To enable face recognition, open the /opt/ffserver/configs/video-manager.yaml configuration file and make sure it contains the face section in detectors that looks similar to the example below.

sudo vi /opt/ffserver/configs/video-manager.yaml

detectors:
  face:
    filter_min_quality: 0.45
    filter_min_size: 1
    filter_max_size: 8192
    roi: ""
    fullframe_crop_rot: false
    fullframe_use_png: false
    jpeg_quality: 95
    overall_only: false
    realtime_post_first_immediately: false
    realtime_post_interval: 1
    realtime_post_every_interval: false
    track_interpolate_bboxes: true
    track_miss_interval: 1
    track_overlap_threshold: 0.25
    track_max_duration_frames: 0
    track_send_history: false
    post_best_track_frame: true
    post_best_track_normalize: true
    post_first_track_frame: false
    post_last_track_frame: false
    tracker_type: simple_iou
    track_deep_sort_matching_threshold: 0.65
    track_deep_sort_filter_unconfirmed_tracks: true
    track_object_is_principal: false
    track_history_active_track_miss_interval: 0
    filter_track_min_duration_frames: 1
    extractors_track_triggers: {}