使用谷歌视觉的CROP_HINTS

Question

我正在使用谷歌视觉API来检测人脸和裁剪相应的图像。这是我的代码，以获得裁剪的coordinates.but，它返回我所拥有的位图图像的最大大小。

                Vision.Builder builder = new Vision.Builder(httpTransport, jsonFactory, null);
                builder.setVisionRequestInitializer(requestInitializer);

                Vision vision = builder.build();

                BatchAnnotateImagesRequest batchAnnotateImagesRequest =
                        new BatchAnnotateImagesRequest();
                batchAnnotateImagesRequest.setRequests(new ArrayList<AnnotateImageRequest>() {{
                    AnnotateImageRequest annotateImageRequest = new AnnotateImageRequest();

                    Image base64EncodedImage = new Image();
                    ByteArrayOutputStream byteArrayOutputStream = new ByteArrayOutputStream();
                    bitmap.compress(Bitmap.CompressFormat.JPEG, 90, byteArrayOutputStream);
                    byte[] imageBytes = byteArrayOutputStream.toByteArray();

                    base64EncodedImage.encodeContent(imageBytes);
                    annotateImageRequest.setImage(base64EncodedImage);

                    ArrayList<Feature> features = new ArrayList<>();
                      Feature cropHints = new Feature();
                    cropHints.setType("CROP_HINTS");
                    features.add(cropHints);
                    annotateImageRequest.setFeatures(features);


                    add(annotateImageRequest);
                }});

                Vision.Images.Annotate annotateRequest =
                        vision.images().annotate(batchAnnotateImagesRequest);

                BatchAnnotateImagesResponse response = annotateRequest.execute();
Sting vertices=response.getResponses().get(0).getCropHintsAnnotation().getCropHints().get(0).getBoundingPoly().getVertices().toString())

            } 

顶点的结果

 [{}, {"x":841}, {"x":841,"y":1499}, {"y":1499}]

Answer 1

这里您想要的是FACE_DETECTION特性，而不是CROP_HINTS。尝试在您的API请求中更改它，您将获得图像中人脸的裁剪坐标。

然后，您可能需要更改处理响应的方式，但是API请求将提供所需的解决方案。我建议您使用JSON进行这里测试，如：

{
  "requests": [
    {
      "image": {
        "source": {
          "imageUri": "<YOUR_SAMPLE_IMAGE_URL>"
        }
      },
      "features": [
        {
          "type": "FACE_DETECTION"
        }
      ]
    }
  ]
}