| 1 | |||
| 2 | package Paws::Rekognition::IndexFacesResponse; | ||
| 3 | use Moose; | ||
| 4 | has FaceModelVersion => (is => 'ro', isa => 'Str'); | ||
| 5 | has FaceRecords => (is => 'ro', isa => 'ArrayRef[Paws::Rekognition::FaceRecord]'); | ||
| 6 | has OrientationCorrection => (is => 'ro', isa => 'Str'); | ||
| 7 | has UnindexedFaces => (is => 'ro', isa => 'ArrayRef[Paws::Rekognition::UnindexedFace]'); | ||
| 8 | |||
| 9 | has _request_id => (is => 'ro', isa => 'Str'); | ||
| 10 | |||
| 11 | ### main pod documentation begin ### | ||
| 12 | |||
| 13 | =head1 NAME | ||
| 14 | |||
| 15 | Paws::Rekognition::IndexFacesResponse | ||
| 16 | |||
| 17 | =head1 ATTRIBUTES | ||
| 18 | |||
| 19 | |||
| 20 | =head2 FaceModelVersion => Str | ||
| 21 | |||
| 22 | The version number of the face detection model that's associated with | ||
| 23 | the input collection (C<CollectionId>). | ||
| 24 | |||
| 25 | |||
| 26 | =head2 FaceRecords => ArrayRef[L<Paws::Rekognition::FaceRecord>] | ||
| 27 | |||
| 28 | An array of faces detected and added to the collection. For more | ||
| 29 | information, see Searching Faces in a Collection in the Amazon | ||
| 30 | Rekognition Developer Guide. | ||
| 31 | |||
| 32 | |||
| 33 | =head2 OrientationCorrection => Str | ||
| 34 | |||
| 35 | If your collection is associated with a face detection model that's | ||
| 36 | later than version 3.0, the value of C<OrientationCorrection> is always | ||
| 37 | null and no orientation information is returned. | ||
| 38 | |||
| 39 | If your collection is associated with a face detection model that's | ||
| 40 | version 3.0 or earlier, the following applies: | ||
| 41 | |||
| 42 | =over | ||
| 43 | |||
| 44 | =item * | ||
| 45 | |||
| 46 | If the input image is in .jpeg format, it might contain exchangeable | ||
| 47 | image file format (Exif) metadata that includes the image's | ||
| 48 | orientation. Amazon Rekognition uses this orientation information to | ||
| 49 | perform image correction - the bounding box coordinates are translated | ||
| 50 | to represent object locations after the orientation information in the | ||
| 51 | Exif metadata is used to correct the image orientation. Images in .png | ||
| 52 | format don't contain Exif metadata. The value of | ||
| 53 | C<OrientationCorrection> is null. | ||
| 54 | |||
| 55 | =item * | ||
| 56 | |||
| 57 | If the image doesn't contain orientation information in its Exif | ||
| 58 | metadata, Amazon Rekognition returns an estimated orientation | ||
| 59 | (ROTATE_0, ROTATE_90, ROTATE_180, ROTATE_270). Amazon Rekognition | ||
| 60 | doesnE<rsquo>t perform image correction for images. The bounding box | ||
| 61 | coordinates aren't translated and represent the object locations before | ||
| 62 | the image is rotated. | ||
| 63 | |||
| 64 | =back | ||
| 65 | |||
| 66 | Bounding box information is returned in the C<FaceRecords> array. You | ||
| 67 | can get the version of the face detection model by calling | ||
| 68 | DescribeCollection. | ||
| 69 | |||
| 70 | Valid values are: C<"ROTATE_0">, C<"ROTATE_90">, C<"ROTATE_180">, C<"ROTATE_270"> | ||
| 71 | =head2 UnindexedFaces => ArrayRef[L<Paws::Rekognition::UnindexedFace>] | ||
| 72 | |||
| 73 | An array of faces that were detected in the image but weren't indexed. | ||
| 74 | They weren't indexed because the quality filter identified them as low | ||
| 75 | quality, or the C<MaxFaces> request parameter filtered them out. To use | ||
| 76 | the quality filter, you specify the C<QualityFilter> request parameter. | ||
| 77 | |||
| 78 | |||
| 79 | =head2 _request_id => Str | ||
| 80 | |||
| 81 | |||
| 82 | =cut | ||
| 83 | |||
| 84 | 1; |