Computer Vision workloads Flashcards
1
Q
Computer vision tasks (capabilities)(largo)
A
Describe images
Tagging visual features
Detect objects
Detect brands
Detect faces
Categorize images
Detect domain-specific contents
(Celebrities, Landmarks)
OCR (Optical character recognition), text in images
Detect image types
Detect image color schemes
Generate thumbnails
Moderate content
2
Q
Computer Vision Solutions
A
Computer vision service Custom vision service Face service Form recognizer Video Indexer
3
Q
Azure Face limitation
A
image no larger than 6MB
formats: JPEG, PNG, GIF (the first frame), BMP
4
Q
Six Common Computer Vision tasks
A
Image classification Object detection Semantic segmentation Image analysis Face detection, analysis, recognition Optical character recognition
5
Q
Image Analysis vs Image Classification
A
Image Classification classify image and videos based on the content. Detect object and text. Extract description and generate tags.
Image Analysis extract info from images, tag them, and create a descriptive image summary.
6
Q
Analyze Image API likelihood two content types
A
Clip art
Line drawing
7
Q
Azure Cognitive Face API service, 8 Face emotion attributes
A
anger contempt disgust fear happiness neutral sadness surprise
8
Q
Custom vision 3 model performance
A
Precision
Recall
Average Precision
9
Q
Object detection model cannot detect objects less than…
A
5% of the image
