1 to 7 of 7 Results
Jan 13, 2023
Bayet, Theophile, 2023, "COCO-style geographically unbiased image dataset for computer vision applications", https://doi.org/10.23708/N2UY4C, DataSuds, V1
There are already a lot of datasets linked to computer vision tasks (Imagenet, MS COCO, Pascal VOC, OpenImages, and numerous others), but they all suffer from important bias. One bias of significance for us is the data origin: most datasets are composed of data coming from develo... |
ZIP Archive - 301.2 KB -
MD5: 7d00b44bfd43f82760b0d1500f661c5e
Annotations for 400 images per geographic zones, JSON format (COCO-style): presence or absence of 91 categories of objects or concepts on the images. |
Adobe PDF - 55.9 KB -
MD5: 808d3cb240f0b589e1ae55e03294151c
Data Management Plan of the project, in French (6 pages) |
ZIP Archive - 168.1 MB -
MD5: b3467eebbefb50907f880809a97fb8df
23 text files: comma separated URLs for each of the 23 geographic zones. Files are named according to geographic zones (UN M49 standard) |
Adobe PDF - 81.4 KB -
MD5: 0e8500cc3f9e2060ccc0bd6ca8c99986
Detailed description of the dataset, in English |
ZIP Archive - 213.9 KB -
MD5: 4288be6f0763234a9becf9bd6f10d2f4
Licenses for the 400 annotations per geographic zones (based on the original licenses specified per image), CSV format. |
Plain Text - 1.7 KB -
MD5: adddf89525947a701c5725394ff712fb
Dataset citation and summary |