v2e Experiment Data README

General information or Introduction section

If you are using this data, please cite:

Y. Hu, S-C. Liu, and T. Delbruck. v2e: From Video Frames to Realistic DVS Events. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), URL: https://arxiv.org/abs/2006.07722, 2021
Y. Hu, T. Delbruck, S-C. Liu, "Learning to Exploit Multiple Vision Modalities by Using Grafted Networks" in The 16th European Conference on Computer Vision (ECCV), Online, 2020.

Contact

Yuhuang Hu
yuhuang.hu@ini.uzh.ch
Institute of Neuroinformatics, University of Zurich and ETH Zurich

Access

The dataset is stored at Zenodo platform, click here to access.

Data and file(s) overview

There are four files

CVPRW_V2E_N_CALTECH_101_DATA.tar.gz contains both training and validation data for reproducing NCaltech-101 experiments in the paper.
mvsec_day_data.tar.gz contains training examples that are taken from the NGA paper (the second item in the citation section).
v2e_day_200hz_all.h5 contains raw v2e events generated using MVSEC dataset's outdoor_day_1 dataset.
v2e_mvsec_train_data.tar.gz contains training data for reproducing MVSEC car detection dataset. For validation, please check out MVSEC-NIGHTL21 dataset.