Skip to content

【活体检测】detectVideo函数总是只返回第一次的检测结果 #17

@MayBlock

Description

@MayBlock

依赖版本:1.0.27
环境: JVM (JDK21)

private fun isLiveFaceByVideo(video: ByteArray): Boolean =
        video.inputStream().use { stream ->
            livenessDetModel.detectVideo(stream).takeIf { it.isSuccess }?.data?.let {
                println("Liveness detection status: ${it.status}, size: ${video.size}")
                it.status == LivenessStatus.LIVE
            }
        } ?: false

detectVideo函数的LivenessStatus总是只返回第一次的检测结果,只有第一次调用正常工作(活体或非活体)
后续即使输入视频或帧发生变化,返回值也始终为第一次的检测结果,疑似存在缓存未失效

活体和人脸检测用的模型都是SEETA_FACE6_MODEL

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions