We identify 1 image over 1 million index under 1 second with only 1 CPU core.
Who we are
LTU is expert in image recognition and artificial vision helping numerous companies detect counterfeiting, measure product popularity, train staff with augmented internal communication, create mobile recognition for consumers. Our clients use our visual intelligence tools we made robust and scalable. LTU’s visual search technologies includes image recognition, fine image comparison, colour search and similarity search based on shape and texture. Our patented algorithms process queries within a second, the best retrieval time available today.
LTU operates on multiple sectors including retail, education, press, media intelligence, brand protection, and law enforcement. LTU provides an API and offers different tools and development kits to ease integration and interfacing to the technology, within the most current software platforms.
Years of research and development led to the creation of algorithms calibrated like the human eye (noise and distortion resistance) combined with processing, indexing and data search.
LTU software and image experts develop and maintain a highly available platform that comes with backoffice applications and integration tools. From providing a simple “off the shelf” account to a full Proof Of Concept in a specific business case, we will work with you to find the best solution and increase your development capacity.
We can also do the full integration regarding the size of your project.
Developing a high availability, reliable and performant image database system learned us how to deploy a robust platform and adress the most needs with a minimum amount of operations. We also provide solutions if you are willing to use Image Recognition with no technical resources in your team.
We are concerned by code quality and always work under unit / functional test coverage. We adhere to the Agile software development principles and put the individuals at the center of the organization.
Our core is written in C++. Engine is made of Python with Django. All of this works on a Linux infrastructure automated with Ansible and Docker/Kubernetes. Peripheral backend engines use NodeJS, Firebase, MongoDB, Elasticsearch. Our main front-end uses BackboneJS. Our progressive web apps run using VueJS. Our mobile SDKs and apps use Swift and Java for Android.
We are concerned by code quality and always work under unit / functional test coverage. We adhere to the Agile software development principles and put the individuals at the center of the organization. Every customer case is considered individually. We follow a method to be the most effective and suit your needs.
We are based in Paris not far from the Eiffel Tower and we're constantly looking for talented people to help us build the most relevant computer vision tools at industrial scale. Join us !