Cognitive service to match an image from an image storage? - microsoft-cognitive

Context: We (a retailer) want to offer our customers the option to upload a picture of a product that they got from our social media accounts, to get all the details about that product (Prices, Stock, description, Variables...etc).
The cognitive service should be able to match and identify which image is this from the image storage.
My question: Do you know a way to use the cognitive services for this use case?

You can use Custom Vision API for your scenario, as it allows you to upload images and label them (on your side), train your model so that you can then evaluate and tag new images (by your users) automatically, and then you can use the images uploaded by your users to improve your classifier.

Related

FlutterFirebase Is it possible to add text data about photo and photo?

I got confused in this situation. Im working on a new project like adding photo and description or other text input. I read a article it is doing this but it's old article.
My app like this; user will add a photo, title and description. Just like a simple Instagram.
(Honestly Im not gonna do instagram. :] )
(Also you can find it here:
https://fireship.io/lessons/flutter-slider-like-reflectly/.)
Is it possible to do that in Firebase Firestore ? Or should i use Cloud Storage?
If I have to use Cloud Storage, how can I add photo description or other text input.
I couldn't find any example like this in Google.
For a user to upload images to your app you would need something like Firebase Storage to host the images, so you will need Cloud Storage anyways. regarding the description, its possible to attach custom metadata to files in Cloud Storage (description, location, title etc...) However if you are going for something more complex, using Firestore together with Storage is probably your best bet. You could store a document for each image in an image collection in firestore and have the imageUrl field point to the Hosting URL along with any other data you might want
You need to use both, you should upload the image in cloud storage and after upload you will get a link of this file. Then you can add this link in firestore document with other information as well such as description or other text input like you wanted.

Is there a way to RESIZE an Image directly in URL using FIREBASE?

I want to crop the Size of an Image directly in the url
I uploaded the images in FireBase and I'm willing to know if they offer this feature.
Any advice and suggestion would be appreciated!
No, this is not a product feature offered by Firebase or Google Cloud products.
Well, firebase does not provide any method to compress or crop images while uploading images. You can crop images before uploading to firebase using:
https://github.com/ivpusic/react-native-image-crop-picker
If you know what sizes you want (maybe your app use predefined image sizes depending on the platform / screen width), you can use this firebase extension (blaze plan).
When you upload an image on storage, it creates multiple resizes on your bucket.
Could be useful depending on your needs.
Edit
For my apps, I wrote a cloud function to upload images on my bucket (React-native expo)
I post my base64 image on an http cloud function and use sharp to resize it.
Then I save my image on my bucket and store its url on firestore (on a document related to this image, a product in my case)

Google Play Store - web scrape top apps for every genre

Currently when I access the Google Play store from a browser (https://play.google.com/store/apps/top?hl=en), I can only see:
Top Apps, Top Selling Apps, Top Grossing Apps, Top Games, Top Selling Games, Top Grossing games.
I am trying to scrape data to get the top selling/grossing apps on the Play Store for EVERY GENRE of apps (i.e. Education, Health, Social, etc). I would've thought this data would be available on the browser Play Store because apparently it is available on the Android App Play Store. When selecting a category on the web browser Play Store, there are no options to view the top selling/grossing apps.
There are many (commercial) API's that give information on top Google Play apps for each genre so surely this information is able to be scraped from somewhere? For example Applyzer, https://www.applyzer.com/?mmenu=worldcharts shows top apps for every genre on the Play store. I would web scrape from here but I want to do this directly from Google.
Any insights would be greatly appreciated.
SOLVED: see my own post below
SOLVED: I just managed to solve this myself. Turns out the Google Play Store does display this information, but you must manually enter the URL yourself. For example: https://play.google.com/store/apps/category/BUSINESS/collection/topselling_paid BUSINESS can be replaced by any category, and topselling_paid can be interchanged with topselling_free or topgrossing

Google API Vision with google cloud storage

I know it's possible to use Google API vision to find similar images on the web.
My goal is to find similar images based on my database of images.
I don't want similar images on the web.
Is it possible?
If yes, can you guide me with some links or some advices?
Thanks!
The Google Vision API web detection feature can be used only to return the contents of the image as they relate to the Google Knowledge Graph, as well as the image's relation to other pages and images on the web.
Based on this, it is not currently possible to use it by specifying your own database of images; However, I suggest you to use the Send Feedback button located at the lower left and upper right corners of the Detecting other features public documentation in order to notify to Google about this desired functionality.
You can upload your images to google cloud and do the job.
You have 2 choices: Google product search or AutoML. With product search you can compare an image with a set of images of products. You will find product form image.
You can also use AutoMl, upload some images and train them. You can predict images with AutoMl API in this case.

Synchronise 2 youtube videos on Drupal site

I want to allow 1 user to point to a youtube video (possibly upload it through the youtube api) on a Drupal site.
And then other users can add videos linked to that video (separate nodes?).
The difficulty is that the videos should be synchronised. So when a user uploads a video to match the original one, he should be able to somehow indicate how this is synchronised with the original one. For instance 4seconds vid1 = 1second vid2
Any ideas, module suggestions, UI ideas are very much appreciated.
I have found this post on video synchronisation with youtube api, but I need a method that would allow the user to actually easily give in the synchronisation. And then synchronise with different starting times.
Best to create a custom module for that. Give the user a form with the youtube id and a sync time then use the youtube api from there. No need for drupal specific code - tho you could create seperate nodes for each video, relate them with node reference fields and store any data in the variables table - butI'd probably create my own table for this.

Resources