Image Stitching Datasets Compilation and Collection

Creating image stitching datasets takes a lot of time and effort. During my Ph.D. days, I tried to compile datasets that were comprehensive to have spherical, cylindrical or planar and full view 360 x 180-degree panoramas. These datasets posed a real challenge to the automatic stitching method. If all these datasets are stitched well, it definitely shows the robustness of your stitching method.

All these datasets are public! Some of them were from my Ph.D. studies (especially on cracks) and most of them were downloaded from the internet. I do not remember the individual names of the dataset providers. But I acknowledge their work and I am thankful to all of them! I hope you appreciate their efforts in making these datasets public to advance the research!

There are 150+ panorama or image stitching/registration datasets in total. Please note that this dataset compilation is more for the qualitative analysis of the image stitching problem. For quantitative analaysis, I recommend using Quantitative Image Stitching Datasets. If I come across any interesting and challenging datasets, I will expand this compilation.

All panorama datasets can be downloaded here Download Dataset. The original GitHub link.

Citation

Image stitching datasets for cracks are available to the public. If you use the dataset related to the cracks in this compilation in your research, please use the following BibTeX entry to cite:

@PhdThesis{preetham2021vision,
	author  = {Aghalaya Manjunatha, Preetham},
	title   = {Vision-Based and Data-Driven Analytical and Experimental Studies into Condition Assessment and Change Detection of Evolving Civil, Mechanical and Aerospace Infrastructures},
	school  = {University of Southern California},
	year    = {2021},
	type    = {Dissertations & Theses},
	address = {3550 Trousdale Parkway Los Angeles, CA 90089},
	month   = {December},
	note    = {Condition assessment, Crack localization, Crack change detection, Synthetic crack generation, Sewer pipe condition assessment, Mechanical systems defect detection and quantification}
}

Acknowledgements

I am thankful to all the authors who made the image stitching datasets public.