Computer Vision for Artists and Designers: Pedagogic Tools and Techniques for Novice Programmers
It was interesting to find out that technological applications stem not only from their technical strength and abilities, but also from suitability of the physical space to which you try to apply computer vision. Therefore, now I can see how in interactive art there are two sided contributions between the artist and computer vision’s capacity that creates a final image that the audience sees.
As it was noted by Golan Levin in the article, computer vision creates broad opportunities for application and interaction with the real world. While it may seem that computer vision almost mimics human vision because of its surveillance practices, at its core there are very different methodologies applied to human and computer vision. While, human vision operates “generally”, analyzing and accepting any image that it sees in the surrounding environment, further generating the same level of processing, computer vision is highly dependent on what kind of image it receives and compatible it is to the vision’s capacity. Therefore, computer vision may generate unexpected processing outcomes with no consistency in the quality. Simply, the way we can help the computer to track what we’re interested in is contrasting the intended object from the background image and other unnecessary surroundings. Just like in Suicide Box project, where vertical motion was characterized as a contrasting event on the image, it was easier for a computer to identify what was tracking target.
In the socially meaningful projects, like Suicide Box, it is confusing to observe how computer vision gets just manually involved in the process of identifying vertical motion without directly understanding what kind of tragic event it gets to record. While, if we would task a specific person to perform the same job as the computer, and see how people commit suicide on the bridge, the person definitely would be left with the trauma, and it would be much harder to transform his knowledge into other forms. While, in this case, using computer vision, we are separating some subjective experiences from the process, and we are left with the intended outcome, which is easier to work on.