A Compact 3D Camera Suited for Mobile and Embedded Vision Applications

Abstract

Recent years have seen the widespread diffusion of 3D sensors, mainly based on active technologies such as structured light and Time-of-Flight, enabling the development of very interesting 3D vision applications. This paper describes a compact 3D camera based on passive stereo vision technology suited for mobile/embedded vision applications. Our 3D camera is very compact, the overall area of the processing unit is smaller than a business card, lightweight, it weights less than 100 g including lenses, has a reduced power consumption, about 2 Watt processing stereo pairs at 30+ fps, and can be easily configured with different baselines and processing units according to specific application requirements. The overall design is mapped on a low cost FPGA, making the hardware design easily portable to other reconfigurable devices, and allows us to obtain in real-time accurate and dense depth maps according to state-of-the-art stereo vision algorithms.

Cite

Text

Mattoccia et al. "A Compact 3D Camera Suited for Mobile and Embedded Vision Applications." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2014. doi:10.1109/CVPRW.2014.36

Markdown

[Mattoccia et al. "A Compact 3D Camera Suited for Mobile and Embedded Vision Applications." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2014.](https://mlanthology.org/cvprw/2014/mattoccia2014cvprw-compact/) doi:10.1109/CVPRW.2014.36

BibTeX

@inproceedings{mattoccia2014cvprw-compact,
  title     = {{A Compact 3D Camera Suited for Mobile and Embedded Vision Applications}},
  author    = {Mattoccia, Stefano and Marchio, Ilario and Casadio, Marco},
  booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
  year      = {2014},
  pages     = {195-196},
  doi       = {10.1109/CVPRW.2014.36},
  url       = {https://mlanthology.org/cvprw/2014/mattoccia2014cvprw-compact/}
}