top of page
ALGORITHM
 Color-Based Segmentation Using K-Means. Automatically choose the right cluster. For now, CPK requires manual edge detection. |  Compares a frame to a background model to determine whether pixels are part of the background or the foreground. The background is acquired in long period of time. By using background subtraction, you can detect foreground objects in an image taken from a stationary camera. |  Center of Mass and distance in Pixels . |
|---|---|---|
 Given an input image I, we compute several channels C = O(I), sum every block of pixels in C, and smooth the resulting lower resolution channels. (3 color, 1 magnitude, 6 gradient) Features are single pixel lookups in the aggregated channels. Boosting is used to learn decision trees over these features (pixels) to distinguish object from background. With the appropriate choice of channels and careful attention to design, ACF achieves state-of-the-art performance in pedestrian detection. |  |
Features
1/3
“However beautiful the strategy,
you should occasionally
look at the results.”
Winston Churchill
Identifies pool edges.
Identifies and tracks moving objects only if human.
Distinguishes between children and adults and between children and animals.
Identifies body progression.
Deals with complex situations, difficult angles, lighting changes and weather.
bottom of page