Output Data Generated for Face Detection, Object Detection, Image Classification, and Text Detection Analysis Models

When you analyze images using an OCI Vision model, the Oracle Analytics data flow generates data about the images. For example, each prediction has a confidence score and a status summary.

Tip: To locate generated datasets, from the Oracle Analytics home page, navigate to Data, then Datasets

Output Data for Images

Face Detection for Images

Output Column	Type	Description
Confidence	Number	Specifies the confidence level of the prediction that the image contains a face.
ID	Number	Specifies a unique ID for each face identified in each image. For example, faces might have IDs 1.00, 2.00, and 3.00.
Image File from Bucket	String	Specifies the object storage URL of the image from the bucket specified. This column is included only when Input type is Bucket.
Left/Right Edge Of Mouth Landmark	String	Specifies the location of the edge of the mouth.
Left/Right Eye Landmark	String	Specifies the location of left and right eyes.
Line Vertex 1, Line Vertex 2, Line Vertex 3, Line Vertex 4	String	Specifies the location (x and y coordinates) of the face.
Nose Tip Landmark	String	Specifies the location of the tip of the nose.
Quality Score	String	Specifies the quality score that is determined by whether the face is clear and unobstructed.
Status Summary	String	Specifies a summary of the model's output, for example, "unsupported image type". This column is blank on successful output.

Image Classification for Images

Output Column	Type	Description
Confidence	Number	Specifies the confidence score of the object detection on the scale of 0 to 1.
Image File from Bucket	String	Specifies the object storage URL of the image from the bucket specified. This column is included only when Input type is Bucket.
Object Name	String	Specifies the identified object's name.
Parents	String	Specifies the categorization of the identified object name.
Status Summary	String	Specifies a brief summary of the model's output, for example, "unsupported image type". This column is blank on successful output.
Synonyms	String	Not used.

Object Detection for Images

Output Column	Type	Description
Confidence	Number	Specifies the confidence score of the object detection on the scale of 0 to 1.
Image File from Bucket	String	Specifies the object storage URL of the image from the bucket specified. This column is included only when Input type is Bucket.
Object Name	String	Specifies the identified object's name.
Parents	String	Specifies the categorization of the identified object name.
Status Summary	String	Specifies the a brief summary of the model's output, for example, "unsupported image type". This column is blank on successful output.
Synonyms	String	Not used.
Vertex 1, Vertex 2, Vertex 3, Vertex 4	String	Specifies the highest points of the highlighted box surrounding an identified object. For example, x_1 = 0.001; y_1 = 0.0013333333333333333, and so on.

Text Detection for Images

Output Column	Type	Description
Image File from Bucket	String	Specifies the object storage URL of the image from the bucket specified. This column is included only when Input type is Bucket.
Line Confidence	Number	Specifies the confidence score of the line of words detected on the scale of 0 to 1.
Line Vertex 1, Line Vertex 2, Line Vertex 3, Line Vertex 4	String	Specifies the vertices (x and y coordinates) of the highlighting box of the line, for example, x_1 = 0.001; y_1 = 0.0013333333333333333.
Line Words	String	Specifies the words detected and extracted as line.
Status Summary	String	Specifies brief summary of the model's output, for example, "unsupported image type". This column is blank on successful output.
Word	String	Specifies the words detected.
Word Confidence	String	Specifies the confidence score of the words detected on the scale of 0 to 1.
Word Vertex 1, Word Vertex 2, Word Vertex 3, Word Vertex 4	String	Specifies the points of a box surrounding an identified word. For example, x_1 = 0.001; y_1 = 0.0013333333333333333.

Output Data for Videos

Face Detection for Videos

Output Column	Type	Description
Frame Confidence	String	Specifies the confidence level of the prediction. For example, 0.97930855.
Frame Time Offset ms	String	For example, 2500.
Frame vertex 1, 2, 3, and 4	String	For example, x_1 = 0.2583038330078125; y_1 = 0.03305579291449653, x_2 = 0.6563463687896729; y_2 = 0.03305579291449653, x_3 = 0.6563463687896729; y_3 = 0.8938908894856771, x_4 = 0.2583038330078125; y_4 = 0.8938908894856771
ID	String	The unique ID of the video, as specified in the input dataset. For example, 1.
Quality Score	String	For example, 0.89911604.
Segment Confidence	String	For example, 0.93692553.
Segment End Time Offset ms	String	For example, 3500.
Segment Start Time Offset ms	String	For example, 2500.
Status Summary	String	A brief summary of the model's output, for example, "unsupported video type". This column is blank on successful output.
URL	String	For example, https://<server>.objectstorage.us-ashburn-1.oci.<customer>-oci.com/n/analyticsdev/b/video_analysis/o/Video_MyVideo.mp4.

Label Detection for Videos

Output Column	Type	Description
Confidence	String	Specifies the confidence level of the prediction. For example, 0.97930855.
ID	String	The unique ID of the video, as specified in the input dataset. For example, 1.
Segment Confidence	String	For example, 0.93692553.
Segment End Time Offset ms	String	For example, 3500.
Segment Start Time Offset ms	String	For example, 2500.
Status Summary	String	A brief summary of the model's output, for example, "unsupported video type". This column is blank on successful output.
URL	String	For example, https://<server>.objectstorage.us-ashburn-1.oci.<customer>-oci.com/n/analyticsdev/b/video_analysis/o/Video_MyVideo.mp4.
Video Label	String	The label identified in the video. For example, "Fashion accessory".

Object Detection for Videos

Output Column	Type	Description
Frame Confidence	String	Specifies the confidence level of the prediction. For example, 0.97930855.
Frame Time Offset ms	String	For example, 0.
Frame vertex 1, 2, 3, and 4	String	For example, x_1 = 0.6046875; y_1 = 0.19166666666666668, x_2 = 0.65859375; y_2 = 0.19166666666666668, x_3 = 0.65859375; y_3 = 0.5708333333333333, x_4 = 0.6046875; y_4 = 0.5708333333333333
ID	String	The unique ID of the video, as specified in the input dataset. For example, 1.
Segment Confidence	String	For example, 0.93692553.
Segment End Time Offset ms	String	For example, 2000.
Segment Start Time Offset ms	String	For example, 0.
Status Summary	String	A brief summary of the model's output, for example, "unsupported video type". This column is blank on successful output.
URL	String	For example, https://<server>.objectstorage.us-ashburn-1.oci.<customer>-oci.com/n/analyticsdev/b/video_analysis/o/Video_MyVideo.mp4.
Video Objects	String	The object identified in the video. For example, a candle.

Text detection for Videos

Output Column	Type	Description
Frame Confidence	String	Specifies the confidence level of the prediction. For example, 0.97930855.
Frame Time Offset ms	String	For example, 0.
Frame vertex 1, 2, 3, and 4	String	For example, x_1 = 0.6046875; y_1 = 0.19166666666666668, x_2 = 0.65859375; y_2 = 0.19166666666666668, x_3 = 0.65859375; y_3 = 0.5708333333333333, x_4 = 0.6046875; y_4 = 0.5708333333333333
ID	String	The unique ID of the video, as specified in the input dataset. For example, 1.
Segment Confidence	String	For example, 0.93692553.
Segment End Time Offset ms	String	For example, 2000.
Segment Start Time Offset ms	String	For example, 0.
Status Summary	String	A brief summary of the model's output, for example, "unsupported video type". This column is blank on successful output.
Video Text	String	The text identified in the video. For example, "Happy Birthday to you".
Video URLs	String	For example, https://<server>.objectstorage.us-ashburn-1.oci.<customer>-oci.com/n/analyticsdev/b/video_analysis/o/Video_MyVideo.mp4.