MediaPipeFaceLandmarkerBase.cs File Reference

Classes
struct	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >.FaceBlendshapeEstimationData< TNet >
	Packed result for one detected face blendshape output. The memory layout matches one row of the packed blendshape result matrix when output_face_blendshapes is enabled. Coefficients corresponds to the MediaPipe Task API face_blendshapes output represented as 52 coefficients. More...
struct	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >.FaceDetectorGraphResult< TNet >
	Bundled outputs of FaceDetectorGraph, corresponding to the upstream DETECTIONS, FACE_RECTS, and EXPANDED_FACE_RECTS streams in face_detector_graph.cc. More...
struct	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >.FaceLandmarkerEstimationData< TNet >
	Packed result for one detected face. The memory layout matches one row produced by BuildPackedOutputMats. NormLandmarks corresponds to the MediaPipe Task API face_landmarks output represented as 478 <see cref="OpenCVForUnity.Extensions.Vec3f"/> values. Face-level raw presence scores are not included; the internal threshold is configured by the MediaPipeFaceLandmarkerBase<OpenCVForUnity.Extensions.Worker.DnnModule.IDnnInferenceNet> constructor. More...
struct	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >.FaceResult< TNet >
	Intermediate result for one face before packing. Downstream subgraph implementations fill each field before packing. More...
struct	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >.FacialTransformationMatrixEstimationData< TNet >
	Packed result for one detected facial transformation matrix. The memory layout matches one row of the packed facial transformation matrix result when output_facial_transformation_matrixes is enabled. RowMajor4x4 corresponds to the MediaPipe face geometry output stored as a row-major 4x4 matrix in the same ordering as MatrixData. More...
class	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >
	Shared MediaPipe FaceLandmarker algorithm implementation for any IDnnInferenceNet backend. Reproduces the face landmarking graph logic of MediaPipe FaceLandmarker on top of the OpenCV Net DNN module. More...
struct	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase< TNet >.NormalizedRect< TNet >
	Face ROI looped back from the previous frame, corresponding to the upstream NormalizedRect packet. More...

Namespaces
namespace	OpenCVForUnity
namespace	OpenCVForUnity.Extensions
namespace	OpenCVForUnity.Extensions.Worker
namespace	OpenCVForUnity.Extensions.Worker.DnnModule
namespace	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe

Enumerations
enum	OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase.MediaPipeFaceRunningMode : byte
	Execution modes compatible with the MediaPipe FaceLandmarker task. This enum corresponds to the task running mode that switches between per-image processing and stateful video processing. More...

Enumeration Type Documentation

MediaPipeFaceRunningMode

enum OpenCVForUnity.Extensions.Worker.DnnModule.MediaPipe.MediaPipeFaceLandmarkerBase.MediaPipeFaceRunningMode : byte

Execution modes compatible with the MediaPipe FaceLandmarker task. This enum corresponds to the task running mode that switches between per-image processing and stateful video processing.