This Artificial Intelligence Model (AI) Knows How To Detect Novel Objects During Object Detection
In recent years, object detection has become a major task in computer vision. In digital images, the goal is to detect objects such as cars, humans, etc. There are hundreds of methods to answer the question: Where are objects?
The traditional methods attempted to answer this by extracting features such as edges and corners from the image. These methods used a sliding window approach. They checked small portions of the image at different scales, to see if they contained the object. It was a very time-consuming process, and the algorithm could miss the object if there were even a slight change in its shape or lighting.
There was the deep-learning era. The advent of large datasets and increasing computer hardware capabilities made it possible to take advantage of the advances in deep learning to create a robust and reliable object detection algorithm.