Lectures in course CS-442: Computer vision

Computer Vision: Fundamentals and Applications Log in to Mediaspace to watch this video

Covers the fundamentals of computer vision, illusions, challenges, applications, and history.

Human Vision: From World to Images Log in to Mediaspace to watch this video

Explores human vision from the retina to the visual cortex, covering topics like color vision, sensitivity to different wavelengths, and comparing human and computer vision processes.

Delineation: Techniques and Applications Log in to Mediaspace to watch this video

Explores techniques for delineation, including Hough transform, gradient orientation, and shape detection, emphasizing the importance of combining graph-based techniques and machine learning.

Segmentation: Theory and Algorithms Log in to Mediaspace to watch this video

Covers the theory and algorithms behind image segmentation, focusing on region identification and evaluation.

Segmentation: Techniques and Applications Log in to Mediaspace to watch this video

Explores segmentation techniques, including CNNs and U-Net models, for image recognition and analysis, emphasizing time-saving automated methods.

Texture: Analysis and Classification Log in to Mediaspace to watch this video

Explores the analysis and classification of texture in images, emphasizing the role of Machine Learning techniques like Convolutional Neural Networks.

Shape from Shading and Texture Log in to Mediaspace to watch this video

Covers shape from shading, texture, reflectance maps, and specularities in computer vision.

Shape from Texture Log in to Mediaspace to watch this video

Explores recovering surface orientation from image texture under different projections and the use of deep learning for normal prediction.

Shape from Stereo-1 Log in to Mediaspace to watch this video

Explores geometric stereo, epipolar geometry, correlation methods, occlusions, and real-time implementation challenges in stereo vision systems.

Shape From Stereo-2 Log in to Mediaspace to watch this video

Explores stereo vision concepts such as occlusions, window size impact, multi-view stereo, dynamic shape reconstruction, and graph-based segmentation.

Shape from Contours Log in to Mediaspace to watch this video

Explores shape reconstruction from contours using edge-based stereo matching and visual hull techniques for accurate 3D reconstruction.

Computer Vision: Recap Log in to Mediaspace to watch this video

Recaps key concepts in computer vision, challenges, opportunities, and upcoming exam details.

Image Formation: Basics Log in to Mediaspace to watch this video

Introduces the basics of image formation, covering analog and digital images, camera geometry, coordinate systems, and camera calibration.

Camera Calibration: Basics Log in to Mediaspace to watch this video

Introduces camera calibration basics, including parameters estimation, pinhole model limitations, lens imaging, depth of field, and lens distortions.

From World to Images: Radiometry and Sensors Log in to Mediaspace to watch this video

Covers radiometry, vignetting, sensor arrays, sensing, and sensors in camera imaging.

Edge Detection: Basics Log in to Mediaspace to watch this video

Covers the basics of edge detection, including edge duality, discontinuities, descriptors, gradients, and magnitude measurement.

Edge Detection: Basics and Techniques Log in to Mediaspace to watch this video

Introduces the basics of edge detection, including measuring contrast, gradient images, Fourier interpretation, Gaussian functions, Canny edge detector, and industrial applications.

Edge Detection: Deep Learning Insights Log in to Mediaspace to watch this video

Explores the evolution of edge detection techniques, from Canny to deep learning insights.

Hand Pose Estimation Log in to Mediaspace to watch this video

Covers hand pose estimation, regression techniques, and the evolution of image classification models from LeNet to VGG19.

Delineation: Techniques and Dynamics Log in to Mediaspace to watch this video

Explores techniques for delineating objects in images, including dynamic programming and live wire methods.