Sep 9–12, 2019

Saving Antarctic penguins with deep learning

Ganes Kesari (Gramener)
2:40pm2:50pm Tuesday, September 10, 2019
Location: 230 A

Solutions that can generate accurate estimates of counts are in demand. They come in handy to tally the number of people in a video frame; they can help count animals of an endangered species in the wild; they can be used to estimate the count of any defined shape within a picture. Traditional crowd-counting methods and models that use detection or regression-based approaches fall short in such scenarios. They suffer from problems such as occlusion, nonuniform distribution, perspective distortion, camera angles, and background clutter. They are not robust and often fail with even simple changes to the planned settings.

Deep learning-based crowd-counting solutions offer an excellent recourse to such problems. Cascaded convolutional neural networks (CNNs) use density-based estimations to preserve spatial information and can localize the count in addition to estimating the overall tally. Such neural network architectures capture global and local features and have been drastically improved over the past year to achieve remarkable accuracy.

Ganes Kesari provides a background of crowd counting and shares the pros and cons of some of the approaches. He explores a real-world application in the biodiversity conservation space, how AI helped count penguin populations in Antarctica by using time-lapse pictures from camera traps. You’ll leave with an understanding of the implementation challenges faced and the approach used to address them.

Photo of Ganes Kesari

Ganes Kesari

Gramener

Ganes Kesari is a cofounder and head of analytics at Gramener, where he leads analytics and innovation in data science, advising enterprises on deriving value from data science initiatives and leading applied research in deep learning at Gramener AI Labs. He’s passionate about the confluence of machine learning, information design, and data-driven business leadership and strives to simplify and demystify data science.

  • Intel AI
  • O'Reilly
  • Amazon Web Services
  • IBM Watson
  • Dataiku
  • Dell Technologies
  • Intuit
  • Gamalon
  • H2O.ai
  • Hewlett Packard Enterprise
  • MapR Technologies
  • Sisu Data
  • Intuit

Contact us

confreg@oreilly.com

For conference registration information and customer service

partners@oreilly.com

For more information on community discounts and trade opportunities with O’Reilly conferences

Become a sponsor

For information on exhibiting or sponsoring a conference

pr@oreilly.com

For media/analyst press inquires