100 days ML roadmap to become a Machine Learning engineer.
Face and eye detection in OpenCV using Haar cascades. Also wrote a tutorial on how to do this and published in medium.
Haar cascade is not much powerful since it is only for front side faced faces and it is not based upon neural network or so. But it does a good job for simple applications
https://blog.goodaudience.com/real-time-face-and-eyes-detection-with-opencv-54d9ccfee6a8
Learned about SSD(Single Shot MUltiBox Detection) algorithm. There is another competitor in the filed called YOLO(You Only Look Once) algorithm. SSD is used for object localization, classification and detection.
SSD is a complex algorithm. It is based on CNN. It has rivals R-CNN, Fast R-CNN, Faster R-CNN. It can be difficult to pick up correct algorithm for correct task
Wrote a blog post about SSD alogrithm. Learned foundation of Convolutional Neural Networks. In CNN the image scale gets decreases in each convolution but preserves the features so it can identify objects of any scale
CNN is new for me so it is taking time
#####Link to work: https://medium.com/@rabinpoudyal1995/ssd-single-shot-multi-box-detection-for-real-time-object-detection-5f2a06e33a4a