Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Computer vision and face recognition using python
1. Computer Vision and Face
Recognition Using Python
A Prelude Webinar on www.techgig.com to RACE360
by Ratnakar Pandey
26 Aug 2019
LinkedIn Profile- https://www.linkedin.com/in/ratnakarpandey/
Quora Q&A - https://www.quora.com/profile/Ratnakar-Pandey-RP
Data Science Blog – www.datafai.com
SlideShare- https://www.slideshare.net/RatnakarPandey6
Email ID- rpdatascience@gmail.com
2. What is Computer Vision?
Computer vision is an interdisciplinary
scientific field that deals with how
computers can be made to gain high-
level understanding from digital images
or videos.
From the perspective of engineering, it
seeks to automate tasks that the human
visual system can do.
- Wikipedia
6. So, How does Computer Vision Work?
Source : https://www.youtube.com/watch?v=OcycT1Jwsns
7. Some Popular CV Tools
Coding Packages
• OpenCV- Originally developed by Intel in 1999, open source and free to use. Supports
multiple platforms- C++, Python, Java etc. https://opencv.org/
• Dlib- C++ library http://dlib.net/
• Face_recognition- Python based very simple library for face recognition
https://pypi.org/project/face_recognition/ (Today’s demo)
• TensorFlow- Free and open source developed by Google https://www.tensorflow.org/
CV as a service or API
• Google Cloud Vision AI- https://cloud.google.com/vision/ (Today’s demo)
• Amazon Rekognition- https://aws.amazon.com/rekognition/
• Microsoft Azure- https://azure.microsoft.com/en-in/services/cognitive-services/computer-
vision/
9. Demo- Google Cloud Vision AI
▪ Go to
https://cloud.go
ogle.com/vision/
▪ Upload any
image of your
choice for
deriving
analytics on your
images,
including OCR.
17. Demo – Python based Face
Detection using OpenCV and
face_recognition
18. Tools and Packages Required
https://colab.research.google.com/ https://opencv.org/ https://pypi.org/project/face_recognition/
GPU
* GPU and Python are both available in Google Colab
22. Upload Images to the Google Colab Environment. You are all set to code.
We are uploading two images-
1. obama1.jpg
2. many_faces.jpg
23.
24.
25.
26.
27. That was simple as we only had one face in the picture. How about this?
28.
29.
30. All 4 faces in the image
have been correctly
identified by the
algorithm!!
31. What’s Next?
▪ Meet me and other speakers in Race 360 Emerging Technology conference in
Bangalore on Aug 28th 2019. Watch more demo on Computer Vision and be
part of CV workshops https://race360.in/
▪ Learn more on Data Science and other topics free of cost on my blog.
www.datafai.com
▪ Learn other functionality of OpenCV and face_recoginition packages such as
changing facial features, identify people, match faces, video analysis etc.
https://pypi.org/project/face_recognition/
▪ Participate in Kaggle and other competitions in this area-
https://www.kaggle.com/c/the-nature-conservancy-fisheries-monitoring
32. Thank You!!
LinkedIn Profile- https://www.linkedin.com/in/ratnakarpandey/
Quora Q&A - https://www.quora.com/profile/Ratnakar-Pandey-RP
Data Science Blog – www.datafai.com
SlideShare- https://www.slideshare.net/RatnakarPandey6
Email ID- rpdatascience@gmail.com