FaceNetModelConfig
data class FaceNetModelConfig(val imageSquareSize: Int? = null, val embeddingDim: Int? = null, val normalization: NormalizationMethod = NormalizationMethod.SCALE_ZERO_TO_ONE, val matchThreshold: Float = 0.63f, val passiveLivenessThreshold: Float = 0.5f)
Configuration parameters for FaceNet and MobileFaceNet models.
Constructors
Link copied to clipboard
constructor(imageSquareSize: Int? = null, embeddingDim: Int? = null, normalization: NormalizationMethod = NormalizationMethod.SCALE_ZERO_TO_ONE, matchThreshold: Float = 0.63f, passiveLivenessThreshold: Float = 0.5f)
Properties
Link copied to clipboard
the output embedding dimension (e.g. 512 for FaceNet-512, 128 for MobileFaceNet). If null, inferred from the TFLite output tensor shape at runtime.
Link copied to clipboard
the input image dimension in pixels (e.g. 160 for FaceNet, 112 for MobileFaceNet). If null, inferred from the TFLite input tensor shape at runtime.
Link copied to clipboard
the minimum cosine similarity between reference portrait and camera face embeddings required to consider the identity verified.
Link copied to clipboard
the tensor preprocessing normalization method to apply.
Link copied to clipboard
the minimum passive anti-spoofing real-face probability (in [0.0, 1.0]) required to consider the detected face a live subject rather than a presentation attack (photo/screen).