What stood out to me in the reading is how limited computer vision really is compared to human vision. As humans, we don’t think twice about recognizing objects, adjusting to poor lighting, or making sense of what we see in context. A computer, on the other hand, needs specific rules and conditions to function. It does not actually understand meaning but instead works through pixels, patterns, and features. If something changes in the environment, like lighting or background, the system can easily fail. That made me realize how much of computer vision is not about “seeing” the world the way we do but about narrowing down what the computer is expected to detect.
To make computer vision work, artists and designers often shape the environment so the system has fewer obstacles. This can be done by using clear contrasts, better lighting, or markers that help the camera distinguish what matters. There are also tools like background subtraction and motion tracking that simplify what the computer needs to follow. Hardware choices such as certain cameras, filters, or infrared technology also help in making the vision system more reliable.
In interactive art, I think this creates both opportunities and challenges. On the positive side, computer vision allows art to respond to the presence and movements of people, turning viewers into active participants. It makes installations feel alive and immersive in ways that would not be possible without tracking. At the same time, it carries the same logic as surveillance because the system is always watching and recording behavior. For me, this makes computer vision powerful but also a little unsettling, since it forces us to think about what it means to be observed and how that shapes the experience of art.