Image stitching has been practiced in various computer vision and scientific study areas. Many different image stitching algorithms have been proposed by different research groups in the past, and there are many different image stitching software products available on the market. However, a comparison between different stitching software products and an evaluation of them has not been performed so far. Furthermore, most previous quality assessment approaches have not had an adequate number of performance matrices, while others have suffered from the adverse effects of computational complications. Our objective is to identify the best software for panoramic image stitching. In this paper we measure the robustness of different software products by assessing image quality of a set of stitched images. For the evaluation itself, a varied set of assessment criteria is used, and evaluation is performed over a large range of images captured in different scenarios using differing cameras. Results show that Autostitch performs relatively well for all types of scenes and for all types of dataset.
CITATION STYLE
Sharma, S. K., Jain, K., & Suresh, M. (2019). Quantitative evaluation of panorama softwares. In Lecture Notes in Electrical Engineering (Vol. 500, pp. 543–561). Springer Verlag. https://doi.org/10.1007/978-981-13-0212-1_56
Mendeley helps you to discover research relevant for your work.