Projects per year
The proposed dataset is an addition to the one published here - http://dx.doi.org/10.17632/bzxk2n78s9.1 . It is also aimed at benchmarking the performance of SfM software under varying conditions - image positions, but also used cameras. The dataset is comprised on images of 11 objects with varying shapes, sizes and surface textures. It is divided into three parts: - Objects imaged in a semi-circle only from one side - 9 images per object - Objects imaged in a full circle from only one height - 18 images per object - Objects images in a full circle from multiple heights - 54 images per object Three cameras are used to capture the images - Canon 600D, Canon 6D and Canon 5Ds. All images contain EXIF data with used camera parameters. The images do not come with ground truth. If you require image data, plus ground truth meshes for comparison, please refer to this dataset - http://dx.doi.org/10.17632/bzxk2n78s9.1
|Date made available||10 Aug 2020|
- 3D reconstruction
- 3D meshes
- 1 Article in proceeding
Benchmarking Close-range Structure from Motion 3D Reconstruction Software under Varying Capturing ConditionsNikolov, I. A. & Madsen, C. B., 31 Oct 2016, 6th International Euro-Mediterranean Conference (EuroMed 2016). Springer, Vol. 10058 2016.
Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-reviewFile