Feature extraction involves feature detection, description and matching which is the baseline of many computer vision applications like content based image retrieval, image classification, image recognition, object detection etc. Features detected should have greater repeatability and should be able to derive descriptors out of it that are highly distinctive and robust to changes in scale, orientation, rotation, illumination etc. This paper provides an insight about the performance comparison of the long existing SIFT and SURF descriptors. The evaluation is carried out in an experimental setup of object category detection which uses a SVM classifier to detect the category. © BEIESP.