Published October 1, 2026 | Version 1.0

Datasets of "Real-Time Estimation of High-Resolution Flow Fields and Reduced-Order Coordinates from Event-Based Imaging Velocimetry"

  • 1. Department of Aerospace Engineering, Universidad Carlos III de Madrid
  • 2. Department of Mechanical and Aerospace Engineering, Politecnico di Torino
  • 3. DLR Institute of Propulsion Technology, German Aerospace Center

Description

Pulsed event-based imaging velocimetry (EBIV) of a submerged water jet (100 Hz) and of a channel flow over a square rib (5 kHz), measured with a Prophesee EVK4 event camera. For each flow the record contains the raw event recording (Prophesee .raw), a subset of pseudo-images, and the time-resolved velocity fields analysed in the paper. The velocity fields are 9000 snapshots per flow, each processed in two ways: at low resolution with single-pass cross-correlation, as computed in real time, and at high resolution with offline multi-frame, multi-pass cross-correlation. These paired fields are used to train and test real-time estimators of the high-resolution field from the low-resolution one. See README.md for the file contents, formats, flow parameters and the training/validation/test split. The estimators and the acquisition software are in VIBEstream (https://github.com/LFranceschelli/VIBEstream).

Files

README.md

Files (3.9 GB)

Name Size
md5:50625f85818d404d12e2e2b7cb519eb9
112 Bytes Download
md5:113d0cac624ec55625ae25702f217981
755.3 MB Download
md5:8ff46cd7f0eeaf5608572165e9ee838d
334.5 MB Download
md5:26c30dfbbd220b360a208d71fc27fc45
22.2 MB Download
md5:4c869cd018f7e41cb6b72b3464825bdf
13.6 MB Preview Download
md5:068f3b9d89b141c44ebbfaaf18307cb5
3.1 MB Preview Download
md5:70d5d37859036fc9e6602745081a64f5
112 Bytes Download
md5:63b2bbb9f356a8b04a68fcec5f85d945
39.1 kB Preview Download
md5:5418b4828f432e1cf0e833785161a24c
131.6 MB Download
md5:27a26bae2567903b7391984104bdf0cb
112 Bytes Download
md5:3c3450791256ad74e381ee7daf418e10
1.5 GB Download
md5:a8cc00f61efe16be34e3aa728c13f02d
1.0 GB Download
md5:145265a2809cc92280e053da3125a96f
25.8 MB Download
md5:4a18e690757c532e15f0901c52d792f2
56.6 MB Preview Download
md5:2239cf6a6ce09c3e2f4755864832cf0e
8.9 kB Preview Download

Additional details

Funding

European Commission
NEXTFLOW - Next-generation flow diagnostics for control 949085

Software

Repository URL
https://github.com/LFranceschelli/VIBEstream
Programming language
Python