Slides from my PyConZA 2019 Keynote on "Deep Neural Networks for Video Applications"
Don't be afraid of A.I. ... git clone a relevant function (deep learning model), fine-tune it for your use case if required and use it to build cool things! I also do consulting if you get stuck or need help @@@ numberboost.com :P
"Most CCTV video cameras exist as a sort of time machine for insurance purposes. Deep neural networks make it easy to convert video into actionable data which can be used to trigger real-time anomaly alerts and optimize complex business processes. In addition to commercial applications, deep learning can be used to analyze large amounts of video recorded from the point of view of animals to study complex behavior patterns impossible to otherwise analyze. This talk will present some theory of deep neural networks for video applications as well as academic research and several applied real-world industrial examples, with code examples in python."
Note: links are hard to click in SlideShare but are clickable if you download PDF :)
#deeplearning #machinelearning #deeplearningforvideo #convolutionalneuralnetworks #recurrentneuralnetworks #centroidtracking #objectdetection #deepfakes #poseestimation #videomachinelearning #numberboost
PyConZA 2019 Keynote - Deep Neural Networks for Video Applications
1. DEEP NEURAL
NETWORKS
Alex Conway
alex @ numberboost.com
PYCONZA
Keynote 2019
Neither confidential nor proprietary - please distribute ;)
for
Video Applications
11. ORIGINAL FILM
Rear Window (1954)
PIX2PIX MODEL OUTPUT
Fully Automated
RE-MASTERED BY HAND
Painstakingly
https://hackernoon.com/remastering-classic-films-in-
tensorflow-with-pix2pix-f4d551fa0503
52. NEURAL NETWORKS
Set of connected Neurons
with randomly initialized weights
and non-linear activation functions
connected in a Network
that are optimized (learned)
using training data
to minimize prediction error
58. Inputs outputs
hidden
layer 1
hidden
layer 2
hidden
layer 3
Note: Outputsof one layer are inputsinto the next layer
This (non-convolutional)architecture is called a “multi-layered perceptron”
(DEEP) NEURAL NETWORKS
59. HOW DOES A NEURAL
NETWORK LEARN?
New
weight =
Old
weight
Learning
rate- ( )x
“How much
error increases
whenwe increase
this weight”
80. 80
Zeiler, M.D. and Fergus, R., 2014, September. Visualizing and understanding convolutional
networks. In European conference on computer vision (pp. 818-833).
81. 81
Zeiler, M.D. and Fergus, R., 2014, September. Visualizing and understanding convolutional
networks. In European conference on computer vision (pp. 818-833).
103. Fine-tuning A CNN
To Solve A New Problem
96.3% accuracy in under 2 minutes for
classifying products into categories
(WITH ONLY 3467 TRAINING IMAGES!!1!)
188. Deep Learning Indaba
http://www.deeplearningindaba.com
Jeremy Howard & Rachel Thomas
http://course.fast.ai
Andrej Karpathy’s Class on Computer Vision
http://cs231n.github.io
Richard Socher’s Class on NLP (great RNN resource)
http://web.stanford.edu/class/cs224n/
Keras docs
https://keras.io/
GREAT FREE RESOURCES