Today, Artificial Intelligence, Machine Learning, and almost a generation of work are paying off as AI learns to envisage and classify objects like humans. And it indicates a huge and essential step to AI evolution. Researchers from UCLA Samueli School of Engineering and Stanford researchers have developed a computer program with the name of Computer Vision that can recognize partially seen objects alone. It goes further than the standard program or task scenario that confines normal computer behaviors.
This is a very hottest area of AI research and it is one of the imperatives of all AI functions, in terms of creating AI operational in the physical world. The utter scale of research in this area is pinpointing, just about every key research organization is putting a more of work into this field, and this latest feat is a huge and vital breakthrough. In fact, search engine Google just published information regarding its custom Optical Character Recognition (OCR) engine, AI Lens that can spot a billion of products, utilizing various techniques.
However, the new UCLA-Stanford approach can distinguish an object by perceiving just part of it. This is a classic human attribute, where a human brain will observe something and immediately classify it, despite visual impediments or blocked lines of sight. Humans can also envisage the state of the remaining object, based on these partial sights of objects. The new innovative technology makes an assembly methodology so computers can perform these things, including partial images- chunks are read by the computer; the computer learns how to amass these chunks perfectly to recognize the object; it evaluates other objects around the target object to help portrayal and identification of the object.