I want to create a connection between a folder in Google Drive, where I will upload photos, and Amazon Rekognition to recognize the text in the photos. I have already created the folder, established a connection with Google, and connected AWS, but unfortunately, the whole process is not working.
Between Google Drive’s “Watch files in folder” and the Google Drive module “Download file,” I set a filter for the Mime type as “image,” as shown in the screenshot. In the Google Drive “Download file” setup, I entered the file ID manually, and it is mapped to the “Watch files in folder.” In AWS, I configured the image upload type, but I’m uncertain whether I should choose “image data” or “S3 object.”
Hi!
You can use it is as follows:
- If you select “Image data” as the Image upload type you should be able to map the data from the Google Drive
Download a File module
- Alternatively, if you have uploaded your data into an S3 data store (either through Make or some other way) you can reference it via S3 Bucket and Name
Depending on your Make plan, there’s limits on maximum file size and data transfer volume per month.
If you’re dealing with a LOT of data files or large files from a local filestore, it will be best to upload these directly into S3 rather than processing the files through Make.
Thank you. I have decided to keep settings of Rekognition module as per your screenshot.
However it looks like I have a problem with previous steps. My flow stops on “Watch Files in Folder” module.
Download file module is not processing anything. I have 1 jpg in that folder.
ok, update - I have checked with ChatGPT - apparently it only works when a new file is uploaded to the folder. I tested and the flow worked.
Do you know what is the best to add at the end of this scenario to see Rekognition answer? Can that go to slack?
You can also right-click on the Watch Files in a Folder module and select “Choose where to start”. That will reset to the latest file after each run, but it’s useful for testing purposes when you just want to process the same file each time.
You can add any module you like - Slack, Teams, Gmail - or anything else. You can map the output from the Rekognition modules to the input parameters of any following module. That’s the beauty of Make!
2 Likes
