Performance Analysis of Various Feature Detector and Descriptor for Real-Time Video based Face Tracking

Abstract

This paper presents the performance analysis of various contemporary feature detector and descriptor pair for real time face tracking. These feature detectors/descriptors are mostly used in image matching applications. Some feature detectors/descriptors like STAR, FAST, BRIEF, FREAK, and ORB can also be used for SLAM applications due to their high performance. However using only one of these feature detectors for object tracking may not provide good accuracy due to various challenges in tracking like abrupt change in object motion, non-rigid object structure, change in appearance of object, occlusions in the scene and camera motion. But it can be combined other object tracking algorithm to improve the overall tracking accuracy. In this paper we have measured the tracking speed and accuracy of these feature detectors in real time video for face tracking using parameters like average number of detected key points, average detection time of key-point, frame per second and number of matches using OpenCV.

SfM ToAEval: Trade-off Aware Evaluation of Feature Extraction Algorithms in Structure from Motion

“…Feature extraction algorithms can be designed for the purpose of feature detection only (like AGAST [17], FAST [18], and STAR [19]), feature description only (like BEBLID [20], BRIEF [21], DAISY [22], FREAK [23], LATCH [24], LUCID [25], TEBLID [26], and VGG [27]), or both feature detection and description (like AKAZE [28], BRISK [29], KAZE [30], ORB [31], SIFT [32], and SURF [33]). In this paper, algorithms that were designed for feature detection are combined with algorithms that were designed for feature description.…”

A Hybrid approach for Image Acquisition Methods based on Feature Based Image Registration (FBIR)

“…An algorithm called enhanced affline transformation (EAT) is presented for "non-rigid infrared (IR) and visible (VIS)" image registration [18]. In order to carry out research in real time visual face tracking applications, a number of feature detector and descriptor pairs can be utilized [19]. A novel feature matching scheme is proposed in [20] which is spatially invariant.…”

A hybrid approach to find cloned objects in copy move forged images

“…Finally, we have compiled our conclusion and future work in this area in Section 5. Kulkarni et al (2013), Patel et al (2014) and Yu et al (2015) suggested contrast-based picture coordinating, which is a vital part of numerous PC-based applications. Distinctive calculations are utilised for picture handling like SIFT, speeded up robust features (SURF), oriented features from accelerated segment test (FAST) and rotated binary robust independent elementary feature (BRIEF) (ORB), ORB calculation utilises the oFAST calculation to distinguish the component focuses, which is the FAST administrator that has bearing.…”