Over the past decade, computer vision has been triumphant in a wide variety of domains. This ranges from automatic facial recognition for unlocking smartphones to autonomous driving vehicles and the detection of diseases on X-ray images.
Researchers around the world are working on computer vision AI architectures to keep improving recognition rates. These architectures dwarf one another one by one and serve the sole purpose of recognizing even the most minimal structures on images, based on which they can be classified.
These strong architectures are fully focused on the goal of classifying pixels in images, but cannot effectively process other data structures such as speech, text, or byte information from computer files. This means that the respecti ...