These powerpoint slides have been prepared by Hemanth
Singamsetty working with Shailaja Vats, Vidhya Jayakrishnan, and
Amin Charaniya.
April 5, 5-8pm, BE 330
Some of these files are big. If you have difficulty downloading
them, please contact the authors directly for a copy:
Metric Image Rectification of a 2D Planar Scene
by Alex D'Angelo and Dat Nguyen
Final project (25%), Weekly Homeworks (25%), Exams (25%), Class Presentation
(what did we learn from you)+Motivation/Initiative/Communication/Reading/Effort/Class-Participation:
(what did we learn from you?): 25%
This class will be an exciting blend of theory and practice and is
oriented towards problem solving.
Potential theoretical topics include:
(i) Single and Multiple View Geometry,
(ii) Image-Based Rendering,
(iii) 3D Reconstruction from Images,
(iv) Tracking,
(v) Panoramic Video Mosaicing, and
(iv) Light fields.
Emphasis will be on
(i) fundamental concepts needed to gain a deep understanding of the
subject,
and (ii) state-of-the-art research in the area to prepare you to
contribute
towards research in this emerging area at the confluence of vision and
visualization.
For example, fundamental concepts in geometry (projective, affine,
Euclidean)
and topics such as projective image warping, and plenoptic modeling will
be covered.
Students will be encouraged to read state-of-the-art or survey
papers on topics of their choice and present the same in the class.
In practice,
students will get to use a variety of
sensors including GPS (Global Positioning System), stereo camera,
camcorder, orientation tracker, Gyrochip, and LiDAR to capture location,
orientation, texture, and depth. Please note that this is a unique
opportunity to work with these equipments.
Students may also have access to
a variety of high resolution texture, height and architectural data
related to UCSC campus depending upon their final projects.
Necessary GIS (geographic
information systems) concepts such as coordinate systems etc. will
be discussed in order to work with these equipments and data
effectively.
PREREQUISITES:
Working knowledge of Matlab and/or OpenGL/FLTK(or XFORMS/GLUT) and
C (or C++) is required. Maturity to read technical papers in the area
of vision/graphics is expected. Ability to choose a project (with
assistance/guidance if necessary) and determination to follow through
with dedicated work is also expected.
Prior exposure to CMPS160/CMPS260/CMPS 262/CMPS290B/CE264/EE264
is a plus but not required.
REQUIRED WORK:
Presence in almost all classes is a must.
Participation on topics covered in the class and on final
project topics outside the classroom during office hours or
by appointment is expected and highly encouraged.
Weekly progress reports, five homeworks (bi-weekly approx.),
two examinations (one half-way and other towards the end), one
technical presentation, and one final programming project & report
(with intermediate progress discussed, reported, and demonstrated
half-way through the quarter) will be required.
RELATIONSHIP to OTHER CLASSES:
This syllabus is being offered for the first time. Anybody who has taken
CMPS 290B before can repeat this class for full credit. Expected overlap
between this class and CMPE 264 is expected to be less than 20%. No
overlap is expected with any other class.