In those years, around 2016, interest in products based on computer vision, augmented reality, and virtual reality had grown considerably. Computer vision was already being used for a wide variety of problems in areas such as security, quality analysis, business intelligence, education, and health, and was also one of the key technologies in the evolution of industrial environments toward Industry 4.0, where it played a central role in both the collection and interpretation of information and the visualization of data.
Analysis using computer vision techniques was costly, especially when it was necessary to process large quantities of images with short response times and minimal error rates, as was the case in most solutions integrated into industrial processes. This cost arose both in the algorithm design phase, which required powerful machines to train classifiers or neural networks, and in their subsequent execution, whose computational complexity made implementation on devices with limited resources difficult.
In response to this problem, VIMO proposed a platform with two main objectives: to facilitate the reuse of image processing and machine learning algorithms to create technically more complex solutions, and to provide image analysis capability to any device that could connect to the platform regardless of its resources, paying special attention to the platform's scalability, maintainability, and availability.
The platform provided value especially in scenarios where the cost of developing an in-house solution was high, or where it was necessary to provide image recognition capability to devices with limited resources and power, including the possibility of taking advantage of image capture infrastructure already deployed, such as existing video surveillance systems in a facility.
VIMO was one of the six strategic projects that CTIC launched in 2016 to structure its R&D&I activity around its lines of specialization, forming the technological basis of the center's vision technologies line (augmented reality, virtual reality, and computer vision), derived from its strategy for the period 2015-2018.