2 minute read

CV: what does it mean to "see"?

Aristotle begins the Metaphysics with an observation about the link between our desire for knowledge and our delight in sight.

“All men by nature desire to know. An indication of this is the delight we take in our senses; for even apart from their usefulness they are loved for themselves; and above all others the sense of sight. For not only with a view to action, but even when we are not going to do anything, we prefer seeing (one might say) to everything else. The reason is that this, most of all the senses, makes us know and brings to light many differences between things.”

Artificial Intelligence (AI) attempts to develop computer systems that are able to perform tasks that have previously depended on human intelligence. In other words we are trying to endow computers with a human like ability to seek knowledge. Given the deep connection between knowledge and sight, it is no great surprise that research into Computer Vision is becoming increasingly popular.


Computer Vision is the theory and technology that lays the foundation for building artificial systems that obtain information from images or multi-dimensional data. Just as a child must learn to distinguish shapes and associate meaning from the flood of light, color, and form, a computer must learn to do so from a swath of pixels. This process begins by inputting labeled photos and/videos with outlines of distinct figures. The algorithm then attempts to make sense of the input by identifying patterns. For example, they “see” hundreds of thousands of pictures of cars in all different colors and styles and begin to link the commonalities between them. Eventually the algorithm will be shown a new unlabeled picture and asked to make a guess - Is there a car in this picture, yes or no? As the algorithm is fed more and more labeled data and gets more and more feedback on it’s guesses, it becomes increasingly confident in its ability to “see.”


Next week researchers from all over the world, will come together at the annual Computer Vision and Pattern Recognition (CVPR) Conference to present the latest research in Computer Vision. We are so excited to sponsor this event and to learn more about the latest developments and innovative business applications.


For more information about Computer Vision, check out our recent white paper that addresses 5 of the most common questions regarding video annotation for computer vision projects.

 

Read More