This chapter describes methods to extract and represent biologically inspired keypoints. It highlights the design choices that are not contradictory to the current understanding of the human visual system (HVS). The chapter briefly defines some terminologies for the sake of clarity. It presents an overview of the current understanding of the HVS. The chapter highlights models and operations within the front end of the visual system that can inspire the computer vision community to design algorithms for the detection and encoding of keypoints. It also presents details behind the state-of-the-art algorithms to extract and represent keypoints. The chapter presents a new evaluation framework to get additional insight on the performance of a descriptor. It shows how to reconstruct a keypoint descriptor to qualitatively analyze its behavior. The chapter concludes with a discussion on how to design a better image classifier using the reconstructed descriptors.