Using the Go programming language and Emergent neural simulation software, I replicated the human nervous system's ability to achieve facial detection. The layers of our network correspond to the V1, V4, and Inferotemporal Cortex (IT) regions of the human brain. The model was trained and tested on a dataset of images containing 600+ faces and non-face images, and the results showed that it was able to accurately detect faces with an accuracy of 72%.
Problem
Outcome
The results demonstrate the potential of using Go and Emergent to create neural network models that can accurately detect faces from images. Future steps include refining the edge-detection capabilities to detect facial features from images with varying orientations of direction for more robust detection and recognition capabilities.