Computer Vision is the sum of the processes that aim to recreate artificially the human ability to see and understand a picture or a video.


Deep Learning is a subset of Machine Learning, a method based on statistical calculation algorithms – also known as Artificial Neural Networks (ANN) – that mimic the processes that take place in the human brain.

Machine Learning is in turn a branch of Artificial Intelligence (AI).


Deep Learning is a subset of Machine Learning, which is a branch of Artificial Intelligence



Pikkart offers its know-how in Computer Vision and Deep Learning to help clients in the design and development of advanced solutions based on the newest technologies.

Our team of highly trained technicians is able to use every aspect of Computer Vision, Machine Learning and Deep Learning to develop high-performing software solutions on cloud, on desktop computers, and on mobile devices.




Since the founding of our company in 2014 we have had the opportunity to work on Computer Vision and Deep Learning projects with both Italian clients and major corporations.

In addition, we have authored several publications on international journals, lectured various technical conferences, and worked for years with research centers such as:

ImageLab, Softech, Università di Modena e Reggio Emilia


Image Recognition
We build smart services and software solutions based on Image Recognition and Deep Learning: logo detection, automatic tagging, Structure from Motion and more.
Mobile vision 
We build intelligent apps using Computer Vision and Machine Learning applied to the mobile environment.   
Document analysis – OCR
Document Analysis is the digital conversion of a picture into text using Computer Vision and Deep Learning applications for the automatic recognition of layout and characters.
Medical Imaging
Computer Vision can greatly help medical doctors when analyzing complex images, like ultrasound scans, thanks to Deep Learning algorithms that can be taught to recognize visual patterns of particular diseases. Other critical tasks tackled by Computer Vision and Deep Learning algorithms are: image denoising, 3D reconstruction, organ segmentation and visual categorization.
Video surveillance 
Computer Vision is applied to video surveillance for many purposes, including forensic, crowd management and marketing effectiveness measuring. Thanks to Deep Learning, privacy preserving solutions can also be developed to obtain accurate people counting and crowd monitoring.
Quality Control
Industry 4.0 can greatly benefit from Computer Vision systems specifically tuned to assess quality requirements. Using Deep Learning systems, a Computer Vision solution can measure all the visual properties of a product in real-time to give instant feedback on production quality and flaws statistics.







Last update:  29/05/2018