Topics To learn

--- MUST READ ---

Algorithms

Great introduce to algorithm design. (Best to learn in parallel with SICP)

Nonnegative matrix factorization, tensor decomposition, sparse coding, learning mixture models, matrix completion and inference in graphical models. et.c

This course is an introduction to linear optimization and its extensions emphasizing the underlying mathematical structures, geometrical ideas, algorithms and solutions of practical problems. The topics covered include: formulations, the geometry of linear optimization, duality theory, the simplex method, sensitivity analysis, robust optimization, large scale optimization network flows, solving problems with an exponential number of constraints and the ellipsoid method, interior point methods, semidefinite optimization, solving real world problems problems with computer software, discrete optimization formulations and algorithms.

NERVANA / INTEL

1. Nervana Graph (Intel nGraph) DOC
2. NGraph Github
3. Nervana home page
4. Intel(R) Math Kernel Library for Deep Neural Networks (Intel(R) MKL-DNN)
5. NervanaSystems Neon Github

NVIDIA

Nvidia lib for improve computations.

The NVIDIA Deep Learning Accelerator (NVDLA) is a free and open architecture that promotes a standard way to design deep learning inference accelerators. With its modular architecture, NVDLA is scalable, highly configurable, and designed to simplify integration and portability. The hardware supports a wide range of IoT devices. Delivered as an open source project under the NVIDIA Open NVDLA License, all of the software, hardware, and documentation will be available on GitHub. Contributions are welcome.

Software optimization resources

This class is a hands-on, project-based introduction to building scalable and high-performance software systems. Topics include performance analysis, algorithmic techniques for high performance, instruction-level optimizations, cache and memory hierarchy optimization, parallel programming, and building scalable distributed systems.

ComputerVision NanoDegree 1000

In this program, you’ll learn the underlying math and programming concepts that drive pattern recognition, object and image classification tasks, and object tracking systems. This course will cover the latest in deep learning architectures used in industry, and you’ll combine current computer vision and deep learning techniques to power a variety of applications. With the practical skills you gain in this course, you’ll be able to program your own applications, extract information from any kind of image and spatial data, and solve real-world challenges.

DeepLearning NanoDegree 600

Become an expert in neural networks, and learn to implement them in Keras and TensorFlow. Build convolutional networks for image recognition, recurrent networks for sequence generation, generative adversarial networks for image generation, and more.

ComputerVision Robotics MicroMasters

Computer Vision / CG Python (OLD)

CV Лекции и материалы - Общее:

1. "Введение в компьютерное зрение и глубокое обучение" ВМК МГУ весна 2017
2. "Введение в компьютерное зрение и глубокое обучение" ВМК МГУ 2008
3. Курсы лаборатории компьютерной графики при ВМК МГУ

Gstreamer

1. GStreamer features
2. Live stream with GStreamer примеры для gst-launch
3. Howto
4. GStreamer, RTP and live streaming, примеры для python
5. Коллекции платных кодеков
6. Hello World GStreamer на языке С

Corners

1. Fast Event-based Corner Detection

CPython internals: A ten-hour codewalk through the Python interpreter source code

Functional Programming on python with stepik

Khan Academy courses

1. Precalculus
2. Multivariable Calculus

Backprop

1. Backprop2
2. Backprop video
3. Gradient Descent improving

Pro Tips about Human Face Detection

Here are some documents that provide information on Haar cascades.

• Tutorial face detection;
• https://www.youtube.com/watch?v=88HdqNDQsEk;
• Opencv face detection;
• OBJECT DETECTION : FACE DETECTION USING HAAR CASCADE CLASSFIERS;
• Youtube video;

• Face Detection Tutorial
• Haar Cascade Object Detection Face & Eye
• OpenCV Face Detection
• Face Detection using Haar Cascade Classifiers

More about hyperparameters, these are some great resources:

• Practical recommendations for gradient-based training of deep architectures by Yoshua Bengio

• Deep Learning book - chapter 11.4: Selecting Hyperparameters by Ian Goodfellow, Yoshua Bengio, Aaron Courville

• Neural Networks and Deep Learning book - Chapter 3: How to choose a neural network's hyper-parameters? by Michael Nielsen

• Efficient BackProp (pdf) by Yann LeCun

More specialized sources:

• How to Generate a Good Word Embedding? by Siwei Lai, Kang Liu, Liheng Xu, Jun Zhao
• Systematic evaluation of CNN advances on the ImageNet by Dmytro Mishkin, Nikolay Sergievskiy, Jiri Matas
• Visualizing and Understanding Recurrent Networks by Andrej Karpathy, Justin Johnson, Li Fei-Fei