Image Labeling - Computer Vision - Elastic Search
Budget: $30 – $250 USD
We have pictures of thousands of people on our website.
We would like to label our images with computer vision to ingrate:
(1) an elastic search component with labels for pictures,
(2) interacting with users when they upload images (e.g., warning if selfie or off-topic or feedback),
(3) content moderation.
(4) more image-related features related to our business (model photos) could be implemented with OpenCV or similar for example
Background: We are a modeling portal (so image features should be related to modeling)
First, we would need a script (ideally in PHP) doing calls to Amazon API and storing JSON responses in a database structure useful for the above purposes. (We already have a Python script connected to AWS Rekognition to get all Rekognition responses, but this was just an internal quick & dirty script to gain better understanding.) This would also include the hierarchical structure of the labels with parent labels etc. We would need someone to tell us whether a direct API call has any disadvantages compared to storing content in S3 bucket for our purposes. Also we cannot grant server access but this should not be necessary.
Please let me know if this is something you can do and if yes what the budget would be.
Whoever we work with can expect longer-term collaboration on this project.
We could then do the same for Google Cloud vision or other solutions you would recommend to enrich our labeling for some purposes.
Thank you! :)
We would like to label our images with computer vision to ingrate:
(1) an elastic search component with labels for pictures,
(2) interacting with users when they upload images (e.g., warning if selfie or off-topic or feedback),
(3) content moderation.
(4) more image-related features related to our business (model photos) could be implemented with OpenCV or similar for example
Background: We are a modeling portal (so image features should be related to modeling)
First, we would need a script (ideally in PHP) doing calls to Amazon API and storing JSON responses in a database structure useful for the above purposes. (We already have a Python script connected to AWS Rekognition to get all Rekognition responses, but this was just an internal quick & dirty script to gain better understanding.) This would also include the hierarchical structure of the labels with parent labels etc. We would need someone to tell us whether a direct API call has any disadvantages compared to storing content in S3 bucket for our purposes. Also we cannot grant server access but this should not be necessary.
Please let me know if this is something you can do and if yes what the budget would be.
Whoever we work with can expect longer-term collaboration on this project.
We could then do the same for Google Cloud vision or other solutions you would recommend to enrich our labeling for some purposes.
Thank you! :)