Gwênlyn
Screen me not, again
At the risk of boring you again, here's another example of achieving computationally what we see, by combining different takes of what the camera sees.
Upper left is the mosquito screen with the blurred foliage behind it. Upper right is the foliage, slightly distorted by the mosquito screen, which is now so blurred it's essentially invisible.
The lower panel shows the result of combining both images using the "min" function. This means that for each pixel in the image, the color of the darker one between the two source images is used. (Last time I used the "max" because the screen was sunlit.)
The composite image definitely looks like what I see when looking through that mosquito screen. By a combination of stereovision and memory, we essentially focus on both planes (screen and foliage), something the camera cannot achieve.
Screen me not, again
At the risk of boring you again, here's another example of achieving computationally what we see, by combining different takes of what the camera sees.
Upper left is the mosquito screen with the blurred foliage behind it. Upper right is the foliage, slightly distorted by the mosquito screen, which is now so blurred it's essentially invisible.
The lower panel shows the result of combining both images using the "min" function. This means that for each pixel in the image, the color of the darker one between the two source images is used. (Last time I used the "max" because the screen was sunlit.)
The composite image definitely looks like what I see when looking through that mosquito screen. By a combination of stereovision and memory, we essentially focus on both planes (screen and foliage), something the camera cannot achieve.