Each of the 20 volunteers participated in one acquisition session. During the recording session, subjects were asked to perform seven different microexpressions.

The visual data includes 126 videos per person (42 per modality) with 2 different light conditions, Natural (N) and Ambient light (A). Each of the seven selected microexpressions was recorded six times per participant: three times under natural lighting and three times under studio lighting.This results in a balanced dataset, with approximately 120 videos per microexpression for each modality.

Event cameras are a new family of vision sensors that differ completely from conventional cameras: instead of capturing images at a fixed rate, they asynchronously measure per-pixel brightness changes and output the event stream as a sequence of tuples [t, x, y, p] that encode the time t, pixel coordinates (x, y) and p the sign of the brightness changes.

The data acquisition took place in a controlled indoor environment with the ambient temperature set to 25°C. To control the lighting conditions during data acquisition, we used two studio lights placed symmetrically on either side of the setup, securing consistent illumination on the face and enhancing the visibility of facial features. The setup included a white wall as a background, a chair positioned at a fixed distance of 0.25 meters from the cameras, and a high desk to guarantee that both cameras were securely positioned, fixed, and aligned during recording.



The database is recorded using 2 cameras and a laptop. The DAVIS346 [1], [9], is used to record events. Its resolution is 346 × 260. For each event [t, x, y, p], x ∈ [0, 345] and y ∈ [0, 259] and the data is saved with the DV software. The RGB and thermal recordings were acquired with the dual sensor, visible and thermal, camera FLIR Duo R developed by FLIR Systems. The visible sensor is a CCD sensor with a pixel resolution of 1920 × 1080. The thermal sensor of this camera is an uncooled VOx microbolometer and has a pixel resolution of 640 × 512. Figure below shows the two cameras used in the database collection.

A download link for the dataset compressed and a password for decrypting the compressed VETEX ZIP files will be provided after receiving the duly signed license agreement. Please fill in the license agreement and send a scanned copy by e-mail (with subject as “VETEX dataset”) at Jean-Luc.Dugelay@eurecom.fr

Any publication using this database must cite the following paper:

@inproceedings{adra2024beyond,
title = {Beyond RGB: Tri-Modal Microexpression Recognition with RGB, Thermal, and Event Data},
author = {Adra, Mira and Mirabet-Herranz, Nelida and Dugelay, Jean-Luc},
booktitle = {International Conference on Pattern Recognition},
year = 2024,
publisher = {IEEE}
}