Open nbiton opened 5 years ago
Cloudinary has addons that we could benefit from to do this. The 2 addons I'm more interested in are Google Auto Tagging and Amazon Rekognition Auto Tagging. [I personally have a preference for Google which seem more advanced than AWS BTW]
The idea is to see if we can use these tools to automatically detect the type of item that the user is trying to add to the Inspection Report.
@kiatlim can you help us think of a UX/flow that would make sense to do that?
We could then maybe leverage BZ keywords ideally to better categorize /standardize categorization of these items in Unee-T.
For me this would be a killer functionality for inspection and inventory report
@nbiton and @kaihendry what do you think?
@franck-boullier Not sure how that would exactly work with the UI, but it sounds like a great idea :+1:
how that would exactly work with the UI
An idea would be to: 1- ask the user to take a picture of the item first. 2- Use the API to tag the picture 3- Ask the user to confirm the tags 4- Once we have all this create the BZ object (a bug) with all the needed elements
That would be sweet!
then, in case the user takes more picture of the same item we won't need to tag these again: we'll rely on the tagging done initially and attach the additional picture to the BZ bug as we are doing today
I think we can reuse the same flow of adding the inventory (with the suggestion of tags after the photo is taken) into the IR. This is so that the flow of adding an inventory is consistent throughout the app.
Under rooms, there is a section for inventory in the room. So, I am guessing that the inventory tab under the IR would refer to general inventory, not in a specific room?
This is another issue branching out of #486 . Implementing this should include:
See https://www.figma.com/file/SgLcXdmBih1JxVq1lupMiPtr/Unee-T-Designs?node-id=458%3A2672 for design reference