Document Type
Dataset
Source Publication Title
IEEE/CVF Winter Conference on Applications of Computer Vision Workshops
DOI
https://doi.org/10.32855/dataset.2026.02.045
Production/Collection Date
University of Texas at Arlington, TX, USA
Depositor
William J. Beksi
Deposit Date
2-19-2026
Data Type
Video
Abstract
UEOF is the first synthetic underwater event-based optical flow dataset derived from physically-based ray-traced RGBD sequences. It was constructed using a modern video-to-event pipeline applied to rendered underwater videos. It consists of realistic event data streams with dense ground-truth flow, depth, and camera motion. The dataset is composed of 12 minutes and 51 seconds of data across 13,714 RGB frames. This results in a total of 4.94 billion events across all scenes. UEOF exhibits a high dynamic range of motion with a mean flow magnitude of 6.1 px and a median of 3.6 px. The motion distribution is heavy-tailed. While a majority of pixels undergo moderate displacement (with a 95th percentile of 21.20 px), the dataset also includes significant high-magnitude flow upwards of 80 px.
Disciplines
Engineering | Robotics
Publication Date
Spring 2-19-2026
Language
English
License
This work is licensed under a Creative Commons Attribution 4.0 International License.
Recommended Citation
Truong, Nick; Karmokar, Pritam P.; and Beksi, William J., "UEOF" (2026). Event-Based Vision. 1.
https://mavmatrix.uta.edu/rvl_ebv_datasets/1