Implementation 3

For the last implementation, I did edge detection but could not finish the part "Perform an edge detection on each frame and average these over N frames. Then add this average to frame N. Edges should be enhanced by this." I asked help from some other classmates but did not get what was needed. In the end I only did Robert's Edge detection. Apologies sir.

Implementation 2

In Implementation 2, I did not quite understand it at first but had help from friends on understanding the concept. Basically what was needed for this implementation is equalizing then averaging base on a specific number of frames and averaging the number of frames then equalizing. The equalizing part was easy because it was already done for image segmentation in Assignment 1. However, the biggest problem for this implementation is that the histogram does not work. Even after spending a lot of time trying to fix this problem.

Implementation 1

For this implementation, I had to review the implementation in Assignment 1 as it was stated in the assignment paper that this implementation would be similar. To create a solution for this implementation, finding out exactly what was needed is key. Asking my lecturer Mr.Rimi for tips helped a lot and I was able to figure out what to do. The explanation given was that, in a video the lighting may change, however the color of a person's shirt should be the same despite the video being darker or lighter. Therefore, the hue was made constant and the saturation and value made changeable. The result was a User Percentile (HSV) which is similar to that of Assignment 1.

Original Video