Visual search is an AI (Artificial Intelligence) powered technology that enables users to sift through millions of items in order for them to find the best visually matching items.
The limitations of the classical search
Imagine watching a movie with your favorite actor in it. He is wearing a unique suit that
makes him look like the king of action and fashion. You instantly decide that you want a suit
just like his, but don’t know how to find it. Googling “Actor’s name suit from movie X” would
probably not give you a link to a shop where you can buy the suit. Before you know it, hours
of search have went by and you haven’t even found the suit in the store. Is this kind of suit,
or any similar suit, even available for sale?
Many people have faced similar or exact scenarios like the one above. Finding the exact item
one has in mind is a different story when it is a struggle to translate a visual idea into words.
“An image is worth a thousand words” is an expression that exists for a reason. According to
an MIT research, people process 90% of information from vision only.
A study conducted by The Intent Lab, a research partnership between Performics and
Northwestern University, revealed that 59% think visual information is more important than
textual information especially when shopping online for clothing or furniture (85%). Overall,
visual information is preferred by at least 50% of survey respondents in all categories except
for electronics, household goods, and wine and spirits. All this evidence suggests we people
need more intuitive search tools when it comes to converting images into search criteria.
Luckily, the solution exists.
Enter Visual Search
Visual search is an AI (Artificial Intelligence) powered technology that enables users to sift
through millions of items in order for them to find the best visually matching items. After the
user gives the query image, the search process takes a few fractions of a second and delivers
only the visually and semantically relevant results.
Research from Accenture has shown that for 84% of retailers, image commerce or visual
search is important or very important to their future success. In addition to that, 56% of
consumers are interested in the technology and 63% would even consider shifting at least
half their purchases to retailers offering it.
Visual search is perfect for shoppers who face two common dilemmas: “I don’t know what I
want, but I’ll know it when I see it” and “I know what I want, but don’t know what it’s called.”
Camera To Search
Fashion Cam, Furniture Cam
based on visually similar items
Where do we excel
History of proven results
Our team has previously worked with multiple clients who were at first skeptical about the
Visual Search. It is completely normal to be hesitant about the new technology, but after the
first results came in, the hesitation went away. Below are results from one of our clients.
Increase in the adviews per session
Increase in the time spent per visit
More responses per visit
One API call is all it takes to use our Visual Search. We develop our solution independent of
your backend code so you don’t have to worry about dependencies. Our solution is written in
a modular way which enables custom integration features. This means you can use
additional search parameters such as location, price or category. It also enables integration
onto arbitrary platforms which include retail sites, chatbot applications, and mobile
Specially optimized models for your data
When it comes to AI technologies, data means almost everything. By carefully analyzing your
datasets, we make sure our AI models are fine tuned for optimal quality and performance.
Below you can see the difference in top results between our specialised model, and the
general model that Alibaba offers.
Additional image processing services
If you need more than Visual Search.
Dominant colors detection
About Velebit AI
Velebit AI is a company that seeks to connect businesses and users with the Artificial
Intelligence technologies. With Velebit AI, clients get to experience the benefits of AI
throughout use cases such as Object Detection, Image Search and Classification, Natural
Language Processing, and many more.
Our team consists of experienced and motivated data scientists, researchers, and developers
coming from physics, mathematics, and computer science backgrounds. We have worked
together on many successful Data Science projects during the last five years, with several of
them being visual and textual searches for online marketplaces and retailers. The joint
experience has formed an environment based on efficient processes, making the product
development much faster and more robust when compared to less cohesive data science
In previous posts from this series, we discussed how we formatted UWSGI and Python logs in JSON format. We still have one important production component left: the Nginx server. This blog post will describe how the Nginx logging module works, and showcase a simple logging configuration where Nginx logger is configured to output JSON logs.
What is the point in collecting logs and metrics if you don’t use them? In this blog post, we will build upon our previous blog post and connect Fluent Bit log collectors to Elasticsearch along with a basic setup and comparison of Kibana and Grafana, tools often used for visualizing logs
There is a great deal of overlap between the fields of statistics and data science, to the point where many definitions of one discipline could just as easily describe the other. While this is true, there are also many differences between them. Why is statistics important and what is its connection to data science? What is data science? What are their similarities and differences? Let’s try to understand it better at least on a basic level without too much going into subtle details.
We build AI for your needs.
Meet our highly experienced team who just loves to build AI and design its surrounding to incorporate it in your business. Find out for your self how much you can benefit from our fair and open approach.