detect_iOS.md

Detecting face using CLOVA face detector

Setting options

ClovaFaceDetector is based on CLOVAEyes SDK. To create a ClovaFaceDetector instance, you need to set ClovaFaceDetectorOption.
With the options, you can automatically configure the CLOVA Eyes pipeline, which is necessary to implement the required features (e.g. face detection, landmark point detection, etc.) for the eKYC service.

Creating instance

You can configure options and create an instance as the following:

let faceOption  = ClovaFaceDetectorOption()
faceOption.stages.append(ClovaPipelineStage(stageType: .detector, filePath: path))
faceOption.stages.append(ClovaPipelineStage(stageType: .landmarker, filePath: path))
faceOption.stages.append(ClovaPipelineStage(stageType: .aligner, filePath: nil))
faceOption.stages.append(ClovaPipelineStage(stageType: .recognizer, filePath: path))
faceOption.stages.append(ClovaPipelineStage(stageType: .maskDetector, filePath: path))
faceDetector = ClovaFaceDetector(option: faceOption)

Detecting face

The face detection feature of the CLOVA face detector SDK detects multiple faces within an image and returns them in ClovaFaceResult format.
To perform face detection, you must create a ClovaFaceDetector instance and call its detectFace() function.
If you wish to check the return format of FaceResult or various information about the detected faces, kindly refer to Processing various information about the face.

let faceResult = faceDetector.detectFace(with image)