Introduction:

A technical review for the highly influential paper, You Only Look Once: Unified Real-Time Object Detection, or commonly coined as YOLO. Jacob Krucinski and I broke down the fundamental components of the paper's revolutionary methodology to our CS7150: Deep Learning classmates at Northeastern University. 

Takeaways:

• When YOLO was published in 2016, it was the first methodology to achieve real-time object detection and classification in video.

• It streamlines previously complex, multi-joint, neural network architectures into one comprehensive Convolutional Neural Network.

• Concatenates pre-trained object detection convolutions with object classification convolutions to detect and classify all at once! ​​​

​

​

Paper