Interface IFaceDetector
- Assembly
- VisioForge.Core.dll
Defines the contract for a face detection component. This interface extends VisioForge.Core.Types.VideoProcessing.IVideoProcessor, indicating that face detection is a form of video processing.
public interface IFaceDetector : IVideoProcessorImplements
Remarks
Implementations of this interface are responsible for analyzing video frames to identify human faces. The results of the detection are communicated through the VisioForge.Core.Types.VideoProcessing.IFaceDetector.FacesDetected event. Face detection algorithms process video frames to locate faces and extract facial features (eyes, nose, mouth). FacesDetected event is raised for each frame where faces are detected, providing CVFace objects with position and landmark data. Implementations may use various detection algorithms: Haar cascades, HOG+SVM, or deep learning models (MTCNN, RetinaFace, YOLOv8). The interface extends IVideoProcessor to enable integration into video processing pipelines with format negotiation and frame processing. Settings configuration (not part of this interface) typically controls detection parameters like minimum face size, confidence threshold, and search mode. Detection performance depends on algorithm choice, hardware capabilities, frame resolution, and configuration settings. Applications should handle FacesDetected events asynchronously to avoid blocking the video processing pipeline. Face detection is the foundation for face recognition, emotion analysis, attention tracking, and AR applications. Implementations must be thread-safe as video frames may be processed concurrently on multiple threads.
FacesDetected
Occurs when one or more faces are detected in a video frame. The event provides a VisioForge.Core.Types.Events.CVFaceDetectedEventArgs object containing details about the detected faces.
event EventHandler<CVFaceDetectedEventArgs> FacesDetected