Image Keywording and Recognition

Everypixel Image Recognition

Use all the power of AI and machine learning to reduce your costs on image recognition and moderation in your apps and products. The Everypixel Image Recognition is a set of pre-trained models available through the API. Thanks to easy algorithms assignment, developers can start using them once they sign up.

Image Keywording

The model recognizes things and people in the image, and then tags it with a set of precise keywords, which serve to categorize and moderate images.
Stock Photography Scoring

A model to estimate the aesthetics and market potential of professional photographs from the point of view of stock photography market experts.
UGC Photography Scoring

A unique model to estimate the aesthetics of users’ photos trained on 347, 000 of Instagram images.
Age recognition

This model understands how old someone looks. It extracts facial features from a photo and gives an accurate estimation of the age.

Pricing

Pay only for what you really need. No monthly fee or imposed package solutions.

Free

for testing in your app or service.

Free for the first 100 operations per day.

Business

pay as you go, cost $ per 1,000 operations

No restrictions.:

UGC Photo Scoring — 1,000 requests for USD 2.0
Other — 1,000 requests for USD 0.6

Enterprise

we can make an advantageous proposal focusing on your needs.

Contact us.

For stock photographers

Use all the power of our neural networks on the Everypixel DAM.

Customized Model

Did not find a suitable solution? The Everypixel Labs specialists will help you build your perfect Model, tuned and trained to your specifications for even better accuracy.

Contact us →

PetaPixel

The stock photography search engine Everypixel has launched a new web app called Everypixel Aesthetics that uses a neural network to tell you how awesome any photo is

– Michael Zhang
Digital Trends

The auto-keywording tool could prove useful, generating a list of keywords from object recognition as well as less concrete terms, like love, happiness, and teamwork.

– Hillary Grigonis
Motherboard Vice

The neural network estimates the aesthetic quality of photos and predicts their commercial (sales) possibilities. To create this algorithm Everypixel's team has asked designers and experienced stock photographers for a help with generating training datasets.

– Liza van Doorn
fstoppers

I think these are good indicators of tech and photography working together.

– Wouter du Toit
DIY photography

Neural networks do keywording pretty accurately. All of the images got a set of suggested tags, most of which could be useful when you’re uploading photos to stock websites (or anywhere else).

– Dunja Djudjic

Use Cases

Categorization

Development of an automated system to choose categories for photos at the stock agency website.
Security

Identification of safety violation on a construction site. The model recognizes people who are not wearing hard hats.
Choosing the Best Shot

Automatic choice of a portrait photograph as a user profile picture if images are bulk uploaded.
Image Moderation

Automatic moderation by image content and its visual aesthetics. For example, it can tag all photo collages or show data about the image attractiveness.
SERP Results Rating by Attractiveness

Our algorithm can be one of the rating criteria of search results including UGC content, if visual aesthetics is an essential characteristic of your resource.
DAM-systems

It is easier to deal with images if the system recognizes what is shown in a photo, tags it, and categorizes by pre-set criteria. It is easier to search for photos, download and use them.

Time to Cut Costs on Image Keywording, Categorization, and Moderation

Everypixel Image Recognition