EC60002: Computer Vision

EC60002
Course name	COMPUTER VISION
Offered by	Electronics & Electrical Communication Engineering
Credits	4
L-T-P	3-1-0
Professor(s)	http://www.ecdept.iitkgp.ac.in/Eece/facultydetails/ece-dsen
Previous Year Grade Distribution
{{{grades}}}
Semester	{{{semester}}}

Syllabus

Concepts taught in class

Color -

Color Photography, Color Representation, Color Matching and Reproduction, Color Coordinate Systems, Color Differences, Color Vision, Color Filter Array, Demosaicing & Deinterlacing, Color Balancing and Gamma, Color Constancy and Retinex

Features -

Local Descriptors: Corner, SIFT, LBP, Steerable Filters

Edge Detection and Linking: LoG, Canny

Image Features: Shape & Texture, DL based Perceptual Features

Motion: Optical Flow, Block Matching, Global Motion, Flownet

Depth: Depth from Structure, DL based Depth, Structure & Depth from Egomotion

Full-reference Quality: SSIM, FSIM, MOVIE index

Processing -

Generic Filters: LMMSE Filter, Order-statistic Filter, Bilateral Filter, Nonlinear Means, Non-local Means

Video Filters: Spatio-temporal Filtering, Blur Reduction

Super-resolution (SR): Splines, Single Image SR, CNN SR, GAN SR

Deconvolution: Unsharp masking, LMMSE & Bayes based Deconvolution, CNN based Deconvolution

Low-light Image Enhancement: Retinex based Contrast Enhancement, Illumination Enhancement, DL based Enhancement

Dehazing: Single Image Dehazing, Prior-based Image Dehazing, DL based Image and Video Dehazing

Decision-making -

Saliency Computation: Image and Video Saliency, DL based Image and Video Saliency

Retargeting and Inpainting: Seam Carving for Image and Video Retargeting, Image Inpainting, DL based Inpainting

Segmentation: Segmentation using Graph Cuts, Mean Shift and Mode Seeking Segmentation, CNN based Semantic Image and Video Segmentation

Object Detection and Recognition: Contrast based Salient Object Detection (SOD), DL based SOD, Video SOD, You Only Look Once (YOLO), Region based CNN (R-CNN) Variants

Categorization and Captioning: Bag of words, EfficientNet on ImageNet, Image Captioning