Computer Vision

Self-study guide for traditional and ML-based computer vision techniques

Theory

Implementations of important computer vision and machine learning concepts.

Computer Vision

• Background Subtraction
• Colorspace
• Features
• Filters
• Geometry
  • Affine transforms
  • Projective transforms
• HOG Features
• Histograms
• Homography
• Hough Transform
• Image Gradients
• K-Means
• Kalman Filter
• Linear algebra
  • Vectors
  • Matrices
• Morphological Operations
• Optical Flow
• Segmentation
• Thresholding

Deep Learning

• Autoencoder
• CNN
• GAN
• VAE

Applied

Solutions to common tasks with popular libraries: OpenCV, PyTorch, Scikit-learn..

• Classification
  • ResNet
  • SqueezeNet
• Object Detection
• Multi-Object Tracking
  • Ball Tracking
  • Player Tracking
• Image Processing
  • Grayscale
• Segmentation
  • FC-DenseNet
  • UNet
• SfM
  • Image Stitching

Problems

Coding problems and solutions. Mostly computer science fundamentals with a slight focus on computer vision.

• Arrays
• Matrix
• HashMap
• Stacks/Queues
• Strings
• Dynamic Programming
• LinkedLists
• Recursion
• Trees

Papers

Notes on interesting computer vision papers.

Setup

Dependencies

• Anaconda 3
• OpenCV 3
• Pytorch >0.2
• Tensorflow
• GPU

Hardware

• Build Your Own Deep Learning Box

Datasets

• Download link for datasets in this repo.

Resources

Courses

• Udacity Intro to Computer Vision
• Berkeley CS184 Cinoyter Graphics and Imaging

Books

• 3D Math Primer for Graphics and Game Development
• Computer Vision: Algorithms and Applications

Papers

• Awesome Computer Vision Papers
• Synthetic Computer Vision Papers

Datasets

• Awesome Computer Vision
• Synthetic Computer Vision Datasets