Computer Vision

Topics

Notes

Linked

Deep Learning Computer Vision
Convolution

Fully connected layers ignore spatial structure and have too many parameters for grid data. Use local weight-sharing filters that exploit translation invariance.

Machine Learning Computer Vision Deep Learning
Convolutional Neural Networks (CNN)

Fully connected networks ignore spatial structure and have too many parameters for images. Use local receptive fields with shared weights for spatial hierarchy.

Convolution Group Equivariant Convolutional Neural Networks
Computer Vision
Harris Corner Detection

How to detect interest points in images? Find locations where intensity changes significantly in all directions.

Computer Vision
Hough Transform

Detecting geometric shapes in noisy images with edge detection alone is fragile. Vote in parameter space to robustly find lines and circles.