Sources:
Digital Image Processing: Rafael Gonzalez and Richard Woods
Modern Computer Vision with PyTorch: V Kishore Ayyadevara and Yeshwanth Reddy
Deep Learning with PyTorch Step-by-Step: A Beginner's Guide: Daniel Voigt Godoy
& PyImageSearch amazing practical contents
slides:
Classic Image Processing (using OpenCV)
Spatial based processing slides
Feature extraction slides
Frequency based processing slides
Morphology slides
Modern Computer Vision (using Pytorch)
Review on deep learning slides
Image classification slides
Transfer learning slides
Object detection slides (comprehensive review of YOLO architectures)
Segmentation slides
Codes
Basics (OpenCV)
loading_saving , add_sub_images, bit_wise_operation , masking_images, histogram_plotting , smoothing , 2d_filtering ,image_gradient
Histogram equalization (OpenCV)
Canny edge detection, hough_circle (openCV)
applying_Canny , circle_detection
feature extraction (OpenCV)
hog, lbp, sift
Neural networks (Pytorch)
Basics of Pytorch [simple linear regression, simple logistic regression]
Transfer Learning using AlexNet
(In class) Lab Notebooks
Lab 1) Lane detection using Hough transforms (solution) [images needed: road.jpg]
Lab 2) Document scanner using contour detection & perspective transformation (solution)[images needed: invoice.jpg]
Lab 3) Barcode detection using morphology operators & contour detection [images needed: barcode.jpg]
Lab 4) Car detection using HOG and SVM [data needed: data.rar]
Lab 5) Image classification using custom dataset [data needed: data.rar]
Lab 6) Multi-Task classification using transfer learning [in google colab]
Lab 7) Traffic signs detection using YOLO in Pytorch
Lab 8) Instance segmentation using pretrained Mask-RCNN in pytorch [images needed: car.jpg]
Lab 9) Training Mask-RCNN on custom dataset [in google colab]
Homeworks
[send your homeworks to cv.kntu2023@gmail.com]
HW2 (deadline: 5th December) [images]
HW3 (deadline: 29th December) [images]
Final Project deadline: 12th February
(This semester will not include oral presentations. Please consolidate your report, presentation slides, and the chosen paper into a single RAR file. Submit the file to cv.kntu2023@gmail.com, using the subject line "CV Final Project )
Note: Reports that are simple translations will not be graded.
Final Project presentation: 13th February