Image and video processing solutions 15-20% Flashcards
Face app capabilities
identify, verify, find similar, group
What are the capabilities of the computer vision app?
Image analysis, Video analysis, Image classification, object detection, facial analysis
What are the different face detection models?
detection_01, detection_02, detection_03
What’s good about detection_02?
It has improved accuracy on small, side-view, and blurry faces, but it won’t return attributes or landmarks
What’s good about detection_03?
Better accuracy on small faces than 02, and will return attributes and landmarks
What computer vision tier supports OCR?
S2
Which document formats does form recognizer prebuilt read?
PDF, word, and excel
What does A1 and A2 mean in the image classification domain?
Improved accuracy
what formats can OCR read api ingest?
jpeg, png, bmp, pdf, and tiff
How many pages can PDFs and TIFFs be for OCR read api?
2000 pages, first 2 pages for free tier
How large can the image files be for the OCR read api?
<500MB, 4MB for the free tier
What are the image dimensions for the OCR read api?
50^2 to 10,000^2
How many print languages can the OCR read api handle?
164
How many handwritten languages can the OCR read api handle?
9
How does calling the OCR read api work?
You call it first and receive an async ID. You then call again for the results
How to set up Conversational language understanding(CLU)?
Using intents as goals and utterances as signals for those goals. Entities provide specification.
What are the three types of entities in a conversational language understanding?
learned entities, list entities, prebuilt entities
single label vs multi label
Single - one item can have one label, multi - one item can have multiple labels
what do you need to prepare for question answering?
A knowledge base to pull from
