Media Content tools for PromoStandards data accuracy
For Suppliers and Distributors in the Promotional Industry
The Problem:
verified Standards issues
- Every Media Content can only be attached to one part while in practice, most of the time,
the
color
is what
differentiate.
Meaning all sizes for color Blue will have the same media content with the same classification. - Allows the supplier to set color name empty(which is OK) but in practice a lot of suppliers doesn't fill it
- Doesn't provide a way to set the bounding box
- Doesn't provide a way to know the size in inches of the bounding box
store
Suppliers
It is time-consuming...
- to classify images in blank/decorated
- to classify images in single part/group
- add the location for the blank image
- add the color to every blank image
- to get the nearest PMS color based on hex for every part
- to generate all the blank images for every color
- to get the bounding box per image/location
- missing art templates
local_shipping Distributors
It is difficult...
- get the blank image to display corresponding to a part
- generate an image based on a logo or text to display as default image in the store
- get the bounding box per image/location needed for design lab + automatic purchase order
List of Features:
check_circle
Already Implemented
- Classify Blank or Decorated ✅
- Classify Single Part or Group ✅
- Detect Bounding Boxes, in /classify endpoint ✅
- API for getting the nearest PMS color based on hex ✅
- Classify color based on image ✅
- Detect colorName based on parts and colors in images (distance between them) ✅
- Remove the background to an image(using AI) ✅
- Guess decoration Bounding box based on decorated and blank images ✅
- Show Bounding Boxes
- Generates our decorated image based on blank image + bounding box ✅
loop In Progress
- Generate missing blank part images(only changing color)
schedule In Future
- Normalize blank images for every color
- Create a manual way to add class types and other fields to media content based on suppliers
information
- It should have event sourcing so that we can revert all changes done by some user
- The user can manually add class types if they belong to a particular group(the account should be paying the highest plan)
- It should have a way to add a new class type based on the supplier information
- When classifying an image(let’s say from PCNA, all the users can have access to this new image information) collectively we are all improving media content information
- The process should be: to add classification via the classifiers or based on suppliers' info, but for every classification, the last who classified, so we will “close” the classification when a human classifies it
- Also provide a way to save/update detected bounding box by location
- Visualizer for media content service
- Image Resizer
Show Bounding Box API
Given an image and Bounding Box in JSON format(optional)
Test Show Bounding Box API
Pricing
At the moment, we are still analyzing the best way to charge for the service.
You can try it for free, but we will limit the number of requests per day.