25-45%

conversion rate uplift

Up to 10%

average order value increase

Up to 5%

revenue per visitor uplift

Up to 98%

search result accuracy

visual style search

VISUAL SIMILARITY SEARCH

They snap it. They find it. They buy it

Hard to describe it? No problem! Let your customers just point and shoot the things they like to instantly find them in your catalog, along with products of a similar style and design.

Our bespoke AI models that power the visual search technology capabilities work across fashion, furniture, dinnerware, fine art, home decor and many other domains.

VISUAL PART IDENTIFICATION

Find a needle in a haystack

Need to find exactly the right thing among thousands that are similar?

Whenever you’re building a fast product reorder system, or part recognition system, our visual search AI models can distinguish the most subtle features and find exactly what the customer is looking for.

visual part identification
visual recommendations

VISUAL RECOMMENDATIONS

Capture every sale with “more like this“ recommendations

Sometimes the right dress doesn’t come in the right size, or the otherwise perfect armchair has that annoying cushion… So when searching for exactly the right product, customers often want to explore and compare similar products.

With “more like this” recommendations, your customers can easily find similar products with respect to fashion or artistic style. Customers  can even look for “dress with similar hem” or “a vase with similar pattern” to focus on product-specific features.

REVERSE VISUAL SEARCH

Convert your lifestyle images into showrooms

Find and tag your products within lifestyle, runway, magazine, and social media pins and influencer images. Allow your customers to seamlessly move from browsing to shopping.

reverse visual search
AI-assisted catalog curation

AI-ASSISTED CATALOG CURATION

Your customers deserve rich and accurate product attributes. Your curators deserve great tools

Richly attributed catalog data makes the search and browsing experience of your customers magical.

However, how often do you see miss-attributed products? Or products with key attributes missing?

Verify and enrich your catalog with hundreds of style, fashion, and decor tags and attributes. Our image recognition algorithms can automatically tag products and verify existing attributes based on product images. Improve the productivity of your curators with AI assistance.

Our clients

Google logo
Apple logo
Paypal logo
macy's brand logo

RETAIL

Neiman Marcus logo
SHIMANO logo
Grandvision logo
macy's brand logo
Lowes logo
Logo of American Eagle

HI-TECH

Google logo
Verizon logo
IAS logo
2k logo
curiositystream brand logo

MANUFACTURING & CPG

Jabil logo
Stanley Black&Decker logo
Levis logo
Boston Scientific logo
Tesla logo

FINANCE & INSURANCE

Paypal logo
SunTrust logo
logo of travelers brand
Raymond James logo
risers logo
Marchmilennan logo

HEALTHCARE

align logo
Rally logo
talix logo
Vertex logo
Merck logo

How our visual search technology works

VISUAL EMBEDDINGS

Find similar images by similar vectors

Visual search AI models represent or “embed” images with their “fingerprints” – series of numbers called vectors that capture the most essential features of an image. Those fingerprint vectors are created in such a way that similar images have similar fingerprints and are geometrically clustered together. This embedding trick allows the search engine to quickly find similar images by their vector representation.

DETECT, EMBED & SEARCH

Full-fledged image recognition pipelines

Our computer vision models perform all tasks required for successful visual search – from object detection, to segmentation, to vector representation and vector search of the target image.

STATE OF THE ART MODELS

Best tools for the job

Our visual search technology utilizes the most recent advancements in computer vision powered by deep learning algorithms. We custom-select, modify, ensemble, train, and fine-tune state-of-the-art visual models for a particular task.

LARGE SCALE DATASETS

Collect and organize domain specific data

We train our models on all available data related to the task – both customer-specific and publicly available. We perform data cleaning and labeling using advanced unsupervised and semi-supervised techniques to build large scale datasets to achieve the best search results.

Accelerate implementation with our visual search blueprint

We created our visual search blueprint based on large scale implementations in the public clouds as well as on-premise for Fortune-1000 companies. We focused on open source and cloud native software, and state-of-the-art deep learning model architectures to enable seamless deployment on any public cloud or private infrastructure. We partner with AWS, Google Cloud, and Microsoft Azure cloud providers to ensure the highest efficiency and best practices.

Accelerate implementation with Grid Dynamics visual search blueprint

Features

  • Accurate results – up to 98 percent item identification accuracy.
  • Advanced similarity – the AI model takes into account fashion, decor, and artistic style.
  • High throughput – battle-tested architecture handling thousands of parallel searches.
  • Low latency – low latency with optimized vectorizers and fast approximate nearest neighbor search.
  • Highly scalable and robust – share-nothing microservices architecture ensures high scalability and resilience.
  • Integrations – data consumption from message queues, databases, or file dumps and REST APIs ensure seamless integration with the rest of the ecosystem.

Technology stack

  • Infrastructure -AWS, GCP, or Microsoft Azure are supported. On-prem solution is available as well. 
  • Deep learning: A choice ofTensorFlow or Pytorch
  • Vector/ANN index: Milvus or Elasticsearch, as well as embedded implementations
  • Data platform: A choice of Apache Spark, Apache Flink, or Apache Beam are the primary choices along with their cloud wrappers.
  • Feature store: Feast is one option, yet many non-specialized databases and EDW solutions will work
  • Infrastructure -AWS, GCP, or Microsoft Azure are supported. On-prem solution is available as well. 
  • Deep learning: A choice ofTensorFlow or Pytorch
  • Vector/ANN index: Milvus or Elasticsearch, as well as embedded implementations
  • Data platform: A choice of Apache Spark, Apache Flink, or Apache Beam are the primary choices along with their cloud wrappers.
  • Feature store: Feast is one option, yet many non-specialized databases and EDW solutions will work

Read more about our visual search case studies

More Search solutions

Semantic vector search

arrow-right

Open source search

arrow-right

Smart autosuggest

arrow-right

Search performance engineering

arrow-right

Replatform from Endeca

arrow-right

Lucidworks solutions

arrow-right

commercetools solutions

arrow-right

Get in touch

Let's connect! How can we reach you?

    Invalid phone format
    Please fill out this field.
    Submitting
    Give a gift of sight to your online commerce

    Thank you!

    It is very important to be in touch with you.
    We will get back to you soon. Have a great day!

    check

    Something went wrong...

    There are possible difficulties with connection or other issues.
    Please try again after some time.

    Retry