Comparative study of feature extraction algorithms for panorama stitching

Abstract

Panorama stitching is a fascinating and rapidly advancing research field. By integrating many photographs that were taken from various angles and viewpoints, with various exposure and color settings, a seamless image is primarily the aim of panorama stitching. This paper investigates the performance of three widely used feature extraction algorithms Speeded-Up Robust Features (SURF), Scale-Invariant Feature Transform (SIFT), and Oriented FAST and Rotated BRIEF (ORB) for panorama stitching. The study compares these algorithms in terms of accuracy, robustness, and speed. Results indicate that while SURF and SIFT produce more accurate and robust results than ORB, they require longer processing time. The study evaluates the approach on a real-world dataset and demonstrates its effectiveness in creating seamless and visually appealing panoramas. This study provides valuable insights into the trade-offs between different feature extraction algorithms and presents a practical solution for panorama stitching applications.