Para ver un resumen de nuestra web en español, haz click aquí.

Multi-sectoral approach

Interested? Let's talk

We work in different sectors, from sports, media, logistics and industry, to health, entertainment, retail, transport and security. If you need a specific solution, do not hesitate to let us know.

Case Studies

Pixelabs has created a custom NLP system for Sercómex Sense, where conversations are transcribed and analyzed, extracting KPIS automatically.

Automatic quality control system for consumer product labels such as detergents, cleaners, shampoos, make-up or adhesives.

We’ve developed a series of computer vision solutions to study the quality of Henkel’s new hair care products (QA process).

Interested in how deep learning and machine vision can be applied to the biomedical field? OCSA facilitates the recognition and analysis of the porcine sperm cell’s motility and morphology.

Quality control system for automotive parts with deep learning and 3D measurements.

We joined the project to collaborate in the development of its intuitive application, which allows citizens to know how to recycle easily.

We’re reshaping new ways to measure and track sponsors’ exposure and advertising in MotoGP™ and FIM CEV Repsol applying computer vision and deep learning.

We measured the sponsorship using our awarded product SyBA. We gathered all the raw data and charts for the Scottish Government and Glasgow City Council.

Computer vision, deep learning and innovation in the Circular Economy.

Technologies

Click on menu below to browse through our technologies and their applications.

rec-objetos

Object Recognition

 

It concerns two main tasks: Object detection, locating objects in an image and indicating their location with a bounding box;  Object classification, predicting the type or class of an object in an image. For example, detection of banners in sports events and the classification of each banner to different brands.

segmentacion

Object Segmentation

 

It provides information about various regions of interest in an image. It classifies each pixel of the image into one of the multiple classes. Semantic segmentation has no information regarding individual objects whereas instance segmentation gives a unique ID to each object. For example, detect irregular scratches in mechanical pieces and measure its surface.

tracking

Object Tracking

 

Aims to analyze videos to identify and track multiple objects, giving them a unique ID and a bounding box to each object for each frame of the video. For example, following porcine sperm cells and analyzing its motility and trajectory.

hiperespectral

Hyperspectral

 

Non-visible light can provide information about the molecular composition of any object and material. This can be implemented in various processes related to quality control and industrial decision making such as food and recycling industry-related processes.

vision3D

3D Vision

 

With technologies such as infrared and visible stereoscopic cameras, or laser profiling sensors, we can understand the 3D information of the world surrounding us.  This allows us to create multiple applications regarding objects measurements, people counting and tracking, or navigation through space.

vision-micro

Microscopic Vision

 

With specific AI algorithms to visually process and understand microscopic images. We can identify objects of interest, segment them and track them across time to help users with their tasks such as quality control or diagnosis.

datos-hist

Historical Records

 

The analysis of time series helps to understand the evolution of the data, detect anomalies or predict data trends. Clustering allows us to group individuals by their similarities.

datos-dec

Data-Driven Decisions

 

Analyze, visualize and understand your data to make real-time decisions. With our dashboards displaying already trained models data, it is possible to see huge amounts of data in a glance.

mod-predictivos

Predictive Models

 

Record all data through time, allow us to predict what’s going to happen in the future, and anticipate issues or events, whether good or bad, to resolve them before they happen.

rec-texto

Text Recognition

 

Detects and recognizes characters and words. It can be used in handwritten text or printed text in a document. The input might be a scanned document, a photo of a document, a scene photo, or from subtitle text superimposed on an image, like a scoreboard in a match. It can be used for multiple purposes such as claims, tables, delivery notes, bills, traffic signals, and any other kind of text.

lenguaje#2

Natural & Structured Language Processing

 

NLP describes the ability of a machine to understand human language, break it down, and understand what it means. It can help to analyze documents and conversations in order to get insights about them in any business domain language. Extract entities, relationships, sentiments, and tone to get better insights from unstructured and structured language.

voz-trnsl#2

Speech to Text, Text to Speech, and Translation

 

Turning speech into text and vice-versa, we find new ways of human device interaction. It makes possible the deployment of cognitive devices. With the add-on of real-time translation, the interaction can reach anyone, anytime, anywhere.

Already trusted by

Certificaciones Pixelabs
Awards
2019 > Winners eDea Salud  Challenge Pfizer Foundation.
2019 > Nominee Tren Lab.
2018 > Premio Mercurio Innovación.
2017 > Premio Cátedra Mejor Emprendedor.
2016 > Emprendedor XXI.
Funding
2022 > Misiones IA TARTAGLIA.
2021 > Misiones CDTI SEPARA.
2020 > Horizon 2020 European Union Funding for Research & Development.
2018 > CDTI Neotec: Sistema para la Evaluación del impacto de marca; SNEO-20171094.
2017 > Línea Enisa.
2016 > Ader Iberaval.
Talent
Universidad Rey Juan Carlos.
Universidad de la Rioja.
Universidad Politécnica de Madrid.
Universidad de Zaragoza.
UNIR.
Partners & Friends
IBM Watson.
IBM Cloud Services.
AWS Cloud Services.
Google Cloud Services.

 TARTAGLIA (con Nº de referencia TSI-100205-2021-11) está enmarcado dentro del programa Misiones de I+D en Inteligencia Artificial de la agenda España Digital 2025 y de la Estrategia Nacional de Inteligencia Artificial, y está financiado por la Unión Europea a través de los fondos NextGenerationEU. Las acciones realizadas se reportan al Ministerio para la Transformación Digital y de la Función Pública (Nº expediente MIA.2021.M02.0005), correspondiente a los fondos del Plan de Recuperación, Transformación y Resiliencia.

logosX2-UE-MIN-PR

El proyecto SEPARA está subvencionado por el CDTI y ha sido apoyado por el Ministerio de Ciencia, Innovación y Universidades.

Logo Ministerio CDTI

Pixelabs S.L ha sido beneficiaria del Fondo Europeo de Desarrollo Regional
cuyo objetivo es mejorar la competitividad de las Pymes y gracias al cual ha puesto en marcha un Plan de Marketing Digital Internacional con el objetivo
de mejorar su posicionamiento online en mercados exteriores durante el año 2020. Para ello ha contado con el apoyo del Programa XPANDE DIGITAL de la Cámara de Comercio de La Rioja.

Programa financiado por el Fondo Europeo de Desarrollo Regional. Una manera de hacer Europa.

Pixelabs logo emblema UE
FEDER - La Rioja Crece con Europa

“Diagnóstico de Alzheimer usando técnicas de IA sobre imagen de fondo de ojo”.

OT1. Potenciar la investigación, el desarrollo tecnológico y la innovación.

IBM Registered Business Partner mark