Question

我正在努力在iOS应用程序中实现人脸识别。我收到了很多标签，例如“眼镜”或“微笑”，但没有看到一个实际的变量告诉我这是一张人脸（以及什么程度的置信度）。

我缺少什么变量，我们如何使用该功能？

Answer 1

我认为您可能没有使用正确的特征类型，因为看来您是在获取标签而不是面部特征。

我建议您查看Detecting Faces和Face Detection Tutorial文档，在其中可以找到详细信息和一些有用的示例，可以将它们用作参考，以进一步了解detecting faces的过程使用 Vision API 。

Answer 2

您可以按照几个步骤从图像中检测人脸。

创建您的URLRequest

func createRequest() -> URLRequest? {
    // Create your request URL
    if let url =  URL(string: "YOUR_API_KEY") {
        var request = URLRequest(url: url)
        request.httpMethod = "POST"
        request.addValue("application/json", forHTTPHeaderField: "Content-Type")
        request.addValue(Bundle.main.bundleIdentifier ?? "", forHTTPHeaderField: "X-  Ios-Bundle-Identifier")
        let jsonRequest = [
            "requests": [
                "features": [
                    [
                        "type": "FACE_DETECTION",
                        "maxResults": 10 //change as per your requirement
                    ]
                ]
            ]
        ]
        let jsonData = try? JSONSerialization.data(withJSONObject: jsonRequest)
        request.httpBody = jsonData
        return request
    }
    return nil
}

在后台线程中运行请求

let task: URLSessionDataTask = URLSession.shared.dataTask(with: request) { (data, response, error) in
    guard let data = data, error == nil else {
        print(error?.localizedDescription ?? "")
        return
    }
    print(data)// Analyze with this data
}
task.resume()

分析数据（如果要更新任何UI组件，请在主线程上进行分析）

    DispatchQueue.main.async(execute: {
            do {
                guard let json =
                    try JSONSerialization.jsonObject(with: data, options: []) as? [String: Any] else { return }
                guard let responses = json["responses"] as? NSArray else { return }
                if responses.count > 0 {
                    guard let response = responses.firstObject as? [String: Any] else { return }
                    guard let faceAnnotations = response["faceAnnotations"] as? NSArray else {
                        print(false, "No face detected, please try another photo.")
                        return
                    }
                    if faceAnnotations.count > 0 {
                        print("Face successfully detected: \(faceAnnotations.count)")
                    }  else {
                        print("No face detected, please try another photo.")
                    }
                } else {
                    print("Error while face detection process, please try again.")
                }
            } catch {
                print("Error while face detection process, please try again.")
            }            
        })

人脸识别变量（google-cloud-vision）

2 个答案: