Machine Learning solutions with MediaPipe

被引：4
作者：
Quinonez, Yadira [1 ]
Lizarraga, Carmen [1 ]
Aguayo, Raquel [1 ]
机构：
[1] Univ Autonoma Sinaloa, Fac Informat Mazatlan, Mazatlan, Sinaloa, Mexico
来源：
2022 11TH INTERNATIONAL CONFERENCE ON SOFTWARE PROCESS IMPROVEMENT, CIMPS | 2022年
关键词：
Artificial Intelligence; Machine Learning; AI platforms; MediaPipe;
D O I：
10.1109/CIMPS57786.2022.10035706
中图分类号：
TP31 [计算机软件];
学科分类号：
081202 ; 0835 ;
摘要：
This document introduces the basic concepts of Artificial Intelligence (AI) and Machine Learning (ML), and some typical applications are mentioned; It also describes some of the most used AI platforms to develop projects using ML algorithms. Specifically, the implementation of Machine Learning solutions is addressed using the MediaPipe framework developed by Google. MediaPipe uses pre-trained models in TensorFlow, OpenCV to manipulate video, and FFmpeg to handle audio data; in addition, it is available for Android, iOS, C++, Python, and JavaScript.
引用
收藏
页码：212 / 215
页数：4
相关论文
共 15 条
  • [1] From Artificial Intelligence to Explainable Artificial Intelligence in Industry 4.0: A Survey on What, How, and Where
    Ahmed, Imran
    Jeon, Gwanggil
    Piccialli, Francesco
    [J]. IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS, 2022, 18 (08) : 5031 - 5042
  • [2] aws.amazon.com, ABOUT US
  • [3] docs.microsoft, About us
  • [4] google.github, ABOUT US
  • [5] jetbrains, about us
  • [6] Liu Wei, 2021, 2021 IEEE 23rd Int Conf on High Performance Computing & Communications
  • [7] 7th Int Conf on Data Science & Systems
  • [8] 19th Int Conf on Smart City
  • [9] 7th Int Conf on Dependability in Sensor, Cloud & Big Data Systems & Application (HPCC/DSS/SmartCity/DependSys)., P2156, DOI 10.1109/HPCC-DSS-SmartCity-DependSys53884.2021.00321
  • [10] media, US
12