A dataset of images of buildings and landmarks, used for training and testing image-based rendering and view synthesis algorithms.

Dataset

This dataset was created by Alex Lau

Contents

This table enumerates the selected photographs from an initial pool of over 70 images, filtered based on criteria detailed in the discussion of ‘the appropriateness of ground photos’ (see Results and discussion section).

We use the Devil Island dataset to conduct a sensitivity analysis for the number of pixel prompts needed using mean intersection over union (mIoU), difference in perimeter to area ratio (PAR), and area error. An up (down) arrow indicates a measure where a larger (smaller) number is preferred.

Evaluation of final predicted penguin colony areas at Devil Island using mean intersection over union (mIoU), difference in perimeter to area ratio (PAR), area error, and accuracy (i.e. Fig 5 vs. ground truth). 95% confidence intervals are shown. We also show the evaluation of a fully manual approach. An up (down) arrow indicates a measure where a larger (smaller) number is preferred.

This dataset is currently associated with an article that is in the process of being published. Once the publication process is completed, a reference link will be added separately. Until that time, this dataset cannot be used for any academic purposes.

The dataset contains 196.926 images and 10 csv files.

The images derived from "the Image Matching Challenge PhotoTourism 2020 dataset"

https://www.cs.ubc.ca/~kmyi/imw2020/data.html

The csv files obtained from our work to show a comprehensive comparison of well-known conventional feature extractors/descriptors, including SIFT, SURF, BRIEF, ORB, BRISK, KAZE, AKAZE, FREAK, DAISY, FAST, and STAR.

Just for gaussian blur there is another file to see.

The images folder contains the images utilized for this study and derived ones originated from these images. (as a total 196.926 images)

To use results or codes from this study to nite to cite:

Please cite this to use anything from this dataset or codes: ISIK M. 2024. Comprehensive empirical evaluation of feature extractors in computer vision. PeerJ Computer Science 10:e2415 https://doi.org/10.7717/peerj-cs.2415

THE COLUMN NAMES: img-1 and img-2 stands for the compared image names KP stands for keyPoints, goodMatches_normal stands for matching count with Brute Force Matcher GM stands for percentage goodMatches_knn stands for matching count with kNN Matcher img-1-D-time shows duration time for keyPoints extraction for img-1 img-2-D-time shows duration time for keyPoints extraction for img-2 (compared one) img-1-C-time shows duration time for comparing keyPoints for img-1 img-2-C-time shows duration time for comparing keyPoints for img-2 (compared one) total-D-time is the total of img-1-D-time and img-2-D-time. total-C-time is the total of img-1-C-time and img-2-C-time. matcher-time_normal stands for time duration for matching process with Brute Force Matcher

matcher-time_knn stands for time duration for matching process with kNN Matcher

More explanation will here soon.

Evaluation of the the Segment Anything Model (SAM) for penguin colony segmentation using mean intersection over union (mIoU), difference in perimeter to area ratio (PAR), area error, and accuracy (i.e. panels a-c in Figs 3 and 4 vs. ground truth). 95% confidence intervals are shown. An up (down) arrow indicates a measure where a larger (smaller) number is preferred.

Data set with tourism potential based on wildlife biomass and diversity for northern Botswana. Data is per 12' grid and includes mean wildlife biomass and number of species per grid for surveys from 1994 to 1999, and from 2001 to 2007.