Corpus ID: 195069430

MediaPipe: A Framework for Building Perception Pipelines

@article{Lugaresi2019MediaPipeAF,
  title={MediaPipe: A Framework for Building Perception Pipelines},
  author={C. Lugaresi and Jiuqiang Tang and H. Nash and Chris McClanahan and Esha Uboweja and M. Hays and Fan Zhang and Chuo-Ling Chang and M. Yong and J. Lee and W. Chang and W. Hua and M. Georg and M. Grundmann},
  journal={ArXiv},
  year={2019},
  volume={abs/1906.08172}
}
  • C. Lugaresi, Jiuqiang Tang, +11 authors M. Grundmann
  • Published 2019
  • Computer Science
  • ArXiv
  • Building applications that perceive the world around them is challenging. A developer needs to (a) select and develop corresponding machine learning algorithms and models, (b) build a series of prototypes and demos, (c) balance resource consumption against the quality of the solutions, and finally (d) identify and mitigate problematic cases. The MediaPipe framework addresses all of these challenges. A developer can use MediaPipe to build prototypes by combining existing perception components… CONTINUE READING
    23 Citations

    Figures and Topics from this paper

    NNStreamer: Efficient and Agile Development of On-Device AI Systems
    • PDF
    MediaPipe Hands: On-device Real-time Hand Tracking
    • 4
    • PDF
    Fingerspelling recognition using synthetic images and deep transfer learning
    Adversarially Robust Frame Sampling with Bounded Irregularities
    • PDF
    Deformable Neural Radiance Fields
    • 2
    • PDF
    End-to-End Multi-Person Audio/Visual Automatic Speech Recognition
    Developing a Lightweight Rock-Paper-Scissors Framework for Human-Robot Collaborative Gaming
    • Highly Influenced
    • PDF

    References

    SHOWING 1-6 OF 6 REFERENCES
    TensorFlow: A system for large-scale machine learning
    • 8,600
    • PDF
    CNTK: Microsoft's Open-Source Deep-Learning Toolkit
    • 279
    Automatic differentiation in PyTorch
    • 7,389
    MXNet: A Flexible and Efficient Machine Learning Library for Heterogeneous Distributed Systems
    • T. Chen, Mu Li, +7 authors Zheng Zhang
    • Computer Science
    • ArXiv
    • 2015
    • 1,487
    • PDF
    Taming heterogeneity - the Ptolemy approach
    • 1,154
    • PDF