Why can’t computers imitate human vision?
- India congratulates China on its election as vice-president of FATF
- MWC 2018: Samsung Galaxy S9 is not fixing what already works well, but packs better cameras
- MWC 2018: Nokia looks at the past and the future, and tries to perfect both
- Worker rights in India:when actions fail words
- Do companies walk the talk on investing in communities?
New Delhi: Even as scientists and companies around the world are making computers capable of recognizing faces and beating grandmasters at chess, there are still fundamental human functions that computers cannot imitate, such as biological vision.
This is why humans can recognize distorted letters presented by websites called Captcha and computers cannot. Captcha is short for Completely Automated Public Turing test to tell Computers and Humans Apart.
In a recent study published in Journal of Neurophysiology, S.P. Arun, an assistant professor at Indian Institute of Science in Bengaluru, and PhD student Ratan Murty have revealed how the brain interprets the 2-dimensional image falling on the retina. Arun and his team have been studying biological vision at the Centre for Neuroscience.
The eye works like a camera so that light enters through the pupil and the lens diverts light onto a screen called the retina. Neurons leaving the retina carry information about the image to the visual areas in the brain. But biological vision can perform several seemingly simple, yet complex tasks.
“The image on the retina contains relevant as well as irrelevant information,” Arun says. “The same object can produce different images because of changes in lighting, size, position and three dimensional rotations and these irrelevant variations have to be factored out by the brain for it to understand that all these images belong to the same object. This computation is performed by neurons in the visual cortex,” he added.
In the study, the two scientists showed recordings from the inferior temporal cortex of the monkey brain which carries out visual object recognition. They found that flashing an image results in neural activity that builds up and drops over a period of time and that during the build-up of the response, neurons are sensitive to irrelevant variations such as changes in the view point of an object. But in the later portion of the response, neurons respond to the same object ignoring irrelevant stimuli. “This shows that neurons in this area perform this important computation dynamically over time”, said Ratan.
Through a series of experiments, the neuroscientists are attempting to understand how the brain visually processes 3-D objects. “Precisely how the brain ignores all the irrelevant variations is a fundamental problem in vision,” Murty said.