Video and Motion Analysis
16-385 Computer Vision (Kris Kitani)
Carnegie Mellon University
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
Guidelines and tips
Prepare for your exams
Study with the several resources on Docsity
Earn points to download
Earn points by helping other students or get them with a premium plan
An overview of the topics covered in the course 16-385 Computer Vision at Carnegie Mellon University. The course covers topics such as optical flow, image filtering, feature detection, object recognition, stereo vision, image alignment, and tracking in video. The document also mentions various techniques used in computer vision such as Hough Transform, Haar-like, HOG, SURF, SIFT, Bag-of-words, K-means, Naive Bayes, SVM, and Deep Convolutional Neural Networks. The course also covers topics such as camera matrix, pose estimation, triangulation, fundamental matrix, epipolar geometry, and reconstruction.
Typology: Lecture notes
2021/2022
1 / 16
This page cannot be seen from the preview
Don't miss anything!
Related documents
Partial preview of the text
Download Computer Vision and Image Analysis and more Lecture notes Optics in PDF only on Docsity!
Video and Motion Analysis
16-385 Computer Vision (Kris Kitani)
Carnegie Mellon University
Optical flow used for feature tracking on a drone
Roadmap
(Where we have been and where we are going)
Image filtering image pyramids
Image gradients Boundaries Hough Transform
Image Manipulation (January)
Object Recognition (February)
Bag-of-words K-means
Nearest Neighbor Naive Bayes^ SVM
Convolutional Neural Networks (February)
Perceptron Gradient Decent
2 view geometry (March)
x = PX
camera matrix
P
pose estimation
X
triangulation
F
fundamental matrix (^) epipolar geometry Reconstruction
Stereo (April)
Block matching Energy minimization
Stereo Rectification
What you will learn next
Computer Vision for Video
(a.k.a., working with sequential images)
Optical Flow (April)
Constant Flow Horn-Schunck
Ix(p 1
) Iy (p 1
Ix(p 2 ) Iy (p 2
Ix(p 25
) Iy (p 25
u
v
It(p 1
It(p 2
It(p 25
min
u, v
X
ij
⇢
E
d
(i, j) + E
s
(i, j)
Tracking in Video (April)
KLT Mean shift
Kalman Filtering SLAM