Unimore logo AImageLab

Baracca Dataset

About the dataset

The recent spread of depth sensors has enabled new methods to automatically estimate anthropometric measurements, in place of manual procedures or expensive 3D scanners. Generally, the use of depth data is limited by the lack of depth-based public datasets containing accurate anthropometric annotations. Therefore, we propose a new dataset, called Baracca, specifically designed for the automotive context, including in-car and outside views. The dataset is multimodal: it has been acquired with synchronized depth, infrared, thermal and RGB cameras in order to deal with the requirements imposed by the automotive context.

Acquisition devices

Considering the requirements imposed by the automotive context, two different cameras have been exploited to collect the data:

  • Pico Zense DCAM710: based on the Time-of-Flight technology, the spatial resolution of the RGB sensor is 1920x1080 pixels, while the infrared/depth sensor has a resolution of 640x480 pixels. The camera is able to acquire valid depth data in the range of 0.2 - 5 meters up to 30 fps.
  • PureThermal 2: this is a camera board equipped with a FLIR Lepton 3.5, a low-resolution (160x120 pixels) thermal radiometric sensor which runs at 9 fps. Since the device is radiometric, it is possible to retrieve the temperature value from the thermal images.


The following set of anthropometric measurements is provided for each participant: height, shoulder width, forearm and arm length, torso width, leg length and eye height from the ground. We also include some soft-biometric traits: age, sex and weight of each subject. 
In addition, we automatically annotate and release the body pose of the subjects for each recorded image. Specifically, we release the the position of 15 skeleton joints in (x, y) image coordinates.


To download the dataset, we require an email address where we'll send a download link to. This will help us to keep in touch in case errors are found or as updates become available.

By downloading the dataset, you declare that you will use it for research and educational purposes only, any commercial use is prohibited.


We sincerely thank all the people who participated in the experiments that led to the creation of this dataset.


We believe in open research and we are happy if you find this data useful. If you use it, please cite our paper.

 title = {Baracca: a Multimodal Dataset for Anthropometric Measurements in Automotive},
 author = {S. Pini, A. D’Eusanio, G. Borghi, R. Vezzani and R. Cucchiara},
 booktitle = {Proceedings of the International joint Conference on Biometrics (IJCB)},
 year = {2020}