- 最新
- 最多得票
- 最多評論
HI Gary - 2/3 of the way down "Non-storage and storage based operations", it says "Non-storage API operations – In these operations, Amazon Rekognition doesn't persist any information. You provide input images and videos, the operation performs the analysis, and returns results, but nothing is saved by Amazon Rekognition."
The API calls you can make for non-storage (and storage, for that matter) are listed >here.<
Image recognition can be from an S3 bucket, or from a local system.
If you are storing your images in S3, there are multiple options for backup on S3, I would recommend reviewing those against your business needs for retention.
Hi, thanks for the answer. I guess my question is more tailored towards the latter. If I create a collection, then index a photo from my local file system, is the photo itself stored anywhere in an S3 bucket behind the scenes or is AWS only storing the metadata for that photo?
相關內容
- AWS 官方已更新 4 個月前
Thanks for the link. That makes sense for the non-storage API operations. In our case we'd be using storage based operations because we want to create collections. So when it says this on the storage based operations page:
"After you create a face collection and store facial feature information for all faces, you can search the collection for face matches"
I assume it's only storing the facial features / template or is it storing the actual photo itself?
Looks like it doesn't keep the photo, rather the template. "The service does not persist actual image bytes. Instead, the underlying detection algorithm first detects the faces in the input image, extracts facial features into a feature vector for each face, and then stores it in the database. Amazon Rekognition uses these feature vectors when performing face matches." https://docs.aws.amazon.com/rekognition/latest/dg/how-it-works-storage-non-storage.html#how-it-works-storage-based