Extracted Features on AmsterTime Dataset

doi: 10.4121/14572644.v4
The doi above is for this specific version of this dataset, which is currently the latest. Newer versions may be published in the future. For a link that will always point to the latest version, please use
doi: 10.4121/14572644
Datacite citation style:
Yildiz, Burak; Khademi, Seyran (2022): Extracted Features on AmsterTime Dataset. Version 4. 4TU.ResearchData. dataset. https://doi.org/10.4121/14572644.v4
Other citation styles (APA, Harvard, MLA, Vancouver, Chicago, IEEE) available at Datacite
choose version:
version 4 - 2022-06-07 (latest)
version 3 - 2022-04-12 version 2 - 2021-06-08 version 1 - 2021-05-12

This collection contains feature files where the features are extracted on AmsterTime dataset using various methods and models such as SIFT, LIFT, and pre-trained VGG-16, ResNets, NetVLAD, AP-GeM and supervisely and self-supervisely trained models. The details of feature extraction procedure and other details can be found on https://github.com/seyrankhademi/AmsterTime.

  • 2021-05-12 first online
  • 2022-06-07 published, posted
Pickle file
TU Delft, Faculty of Electrical Engineering, Mathematics and Computer Science Intelligent Systems


files (14)