3D Form and Motion: Difference between revisions

Revision as of 00:28, 25 March 2013

Mondays at 1:00, Evans 560

3D models from images:

Hartley and Zisserman: Multiple View Geometry (Amazon)
Hoiem and Savarese: 3D object recognition and scene interpretation, book pdf
Hoiem Ph.D. thesis "SEEING THE WORLD BEHIND THE IMAGE" pdf
Cashman and Fitzgibbon: "What Shape are Dolphins? Building 3D Morphable Models from 2D Images" pdf
Snavely and Seitz (2006), "Photo tourism" (aka photosynth), pdf
Seitz google talk on history of 3D computer vision: URL

SLAM:

wiki page with list of refs
Thrun, "Probabilistic algorithms in robotics" pdf
Daniel Cremers work on helicopters/SLAM (TU Munich) URL
Newcombe and Davison, "Live dense reconstruction with a single moving camera work on fast single camera", pdf

Psychophysics:

Nakayama et al. (1995) "Visual Surface Representation" pdf
Glennerster and Fitzgibbon, "View-Based Approaches to Spatial Representation in Human Vision" pdf
Wexler work on depth from self-motion/parallax, reference frames, pdf1, pdf2

Action-perception:

Philipona et al. Neural Computation (2003) pdf
- Follow on papers URL
Philipona et al. NIPS (2003) pdf
- Follow on papers URL
Robotics paper trying to perceive environment based on sensiromotor loop (from Pulkit): pdf

Manifolds:

@@ Line 7: / Line 7: @@
 * Cashman and Fitzgibbon: "What Shape are Dolphins? Building 3D Morphable Models from 2D Images" [http://redwood.berkeley.edu/bruno/3DFM/cashman-fitzgibbon.pdf pdf]
 * Snavely and Seitz (2006), "Photo tourism" (aka photosynth), [http://redwood.berkeley.edu/bruno/3DFM/snavely-seitz.pdf pdf]
+* Seitz google talk on history of 3D computer vision: [https://www.youtube.com/watch?v=kyIzMr917Rc URL]
 '''SLAM:'''