Lecture Schedule

From cs331b Special Topics in 3dRR

Revision as of 00:31, 26 September 2013 by Soohark (Talk | contribs)
Jump to: navigation, search
Lect. Date Topics Presenter Paper/Slides
1 9/23 Introductions slides
2 9/25 Representing and Recognizing Objects
Guest lecturer: Y. Xiang

3D object recognition by aspect graph-based models:

  • Y.Xiang, S.Savarese. "Estimating the aspect layout of object categories". CVPR2012 link
3 9 /30 Representing and Recognizing Objects   3D detection from D-RGB data:
  • Liefeng Bo, Kevin Lai, Xiaofeng Ren, and Dieter Fox "Object Recognition with Hierarchical Kernel Descriptors"
  • L. Bo, X. Ren, and D. Fox. Kernel Descriptors for Visual Recognition. In NIPS, December 2010
  • K. Lai, L. Bo, X. Ren, and D. Fox. A Large-Scale Hierarchical Multi-View RGB-D Object Dataset. In IEEE International Conference on on Robotics and Automation, 2011
4 10/2 Representing and Recognizing Objects Guest lecturer: Dr. M. Stark 3D object recognition by deformable part models:
  • B.Pepik, M.Stark, P.Gehler and B.Schiele, "Teaching 3D geometry to deformable part models". CVPR2012 link
5 10/7 Representing the 3D Space   Large scale 3D reconstruction by structure from motion :
  • Sameer Agarwal, Yasutaka Furukawa, Noah Snavely, Ian Simon, Brian Curless, Steven M. Seitz and Richard Szeliski Communications of the ACM, Vol. 54, No. 10, Pages 105-112, October 2011 link
  • Jan-Michael Frahm, Pierre Georgel, David Gallup, Tim Johnson, Rahul Raguram, Changchang Wu, Yi-Hung Jen, Enrique Dunn, Brian Clipp, Svetlana Lazebnik, Marc Pollefeys, ECCV 2010
6 10/9 Representing the 3D Space   Indoor scene layout reconstruction:
  • David Lee, Takeo Kanadem Matrial Hebert. "3D Scene Analysis". CVPR2009 link
  • A. Schwing and R. Urtasun, Efficient Exact Inference for 3D Indoor Scene Understanding In European Conference on Computer Vision (ECCV), Florence, Italy, October 2012
7 10/14 Understanding Complex Scenes   Joint 3D reconstruction and object detection:
  • D. Hoiem, A.A. Efros, and M. Hebert, "Putting Objects in Perspective", to appear in CVPR 2006.
  • Thinking Inside the Box: Using Appearance Models and Context Based on Room Geometry; V. Hedau, D. Hoiem, and D.A. Forsyth ECCV 2010
8 10/16 Understanding Complex Scenes Guest lecturer: B. Kim 2D/3D CRF models fo understanding:
  • B.Kim, M.Sun ,P.Kohli, S.Savarese, "Relating things and stuff by High-Order Poterntial modeling" ECCV2012

Project Proposal due    
9 10/21 Understanding Complex Scenes Guest lecturer: Y. Bao Semantic structure from motion:
  • S. Bao and S. Savarese, "Semantic Structure from Motion". Proceedings of the IEEE International Conference on Computer Vision and Pattern Recognition, 2011
10 10/23 Understanding Complex Scenes   Tutorial on MCMC:
  • Monocular 3D Scene Modeling and Inference: Understanding Multi-Object Traffic Scenes, Christian Wojek, Stefan Roth, Konrad Schindler, Bernt Schiele
11 10/28 Understanding Complex Scenes   Joint segmentation and reconstruction from D-RGB data (I):
  • Nathan Silberman, Derek Hoiem, Pushmeet Kohli, Rob Fergus "Indoor Segmentation and Support Inference from RGBD Images" ECCV, 2012 link
  • S. Gupta, P. Arbelaez and J. Malik. Perceptual Organization and Recognition of Indoor Scenes from RGBD Images. In CVPR 2013 (oral). link
12 10/30 Understanding Complex Scenes   Joint segmentation and reconstruction from D-RGB data (II):
  • Munoz, D., Bagnell, J.A., Hebert, M.: Stacked hierarchical labeling. In: ECCV. (2010)
  • Co-inference for Multi-modal Scene Analysis Daniel Munoz, J. Andrew (Drew) Bagnell, and Martial Hebert European Conference on Computer Vision (ECCV), October, 2012.
13 11/4 Understanding Human Pose Guest lecturer: Dr. Roland Angst Joint segmentation and reconstruction from D-RGB data (III):
  • Christian Häne, Christopher Zach, Andrea Cohen, Roland Angst, Marc Pollefeys, Joint 3D scene reconstruction and class segmentation link

Project Mid-term Report due     
14 11/6 Understanding Human Pose   Human pose estimation and activity recognition from 2D


  • Lubomir Bourdev, Jitendra Malik,"Poselets: Body Part Detectors Trained Using 3D Human Pose Annotations", ICCV 2009 link
  • Lubomir Bourdev, Subhransu Maji, Thomas Brox, Jitendra Malik,"Detecting People Using Mutually Consistent Poselet Activations", ECCV 2010 link
  • Action Recognition with Exemplar Based 2.5D Graph Matching Bangpeng Yao and Li Fei-Fei, European Conference on Computer Vision (ECCV). Firenze, Italy. October 7-13, 2012.
15 tba Understanding Human Activities   Human pose estimation and activity recognition from D-RGB


  • Tutorial on Randomized decision trees and forests
  • Jamie Shotton , Andrew Fitzgibbon , Mat Cook , Toby Sharp , Mark Finocchio , Richard Moore , Alex Kipman , Andrew Blake, "Real Time human pose recognition in parts from single depth image". CVPR, 2011 link
  • Mining Actionlet Ensemble for Action Recognition with Depth Cameras , Jiang Wang, Zicheng Liu, Ying Wu, Junsong Yuan link
16 11/11 Understanding Human Activities Guest lecturer: Dr. W. Choi  Recognizing collective activities:
  • W. Choi, K. Shahid, S. Savarese "What are they doing?: Collective Activity Classification Using Spatio-Temporal Relationship Among People", 9th International Workshop on Visual Surveillance (VSWS09) in conjuction with ICCV 09 PDF
  • W. Choi and S. Savarese "A Unified Framework for MultiTarget Tracking and Collective Activity Recognition" ECCV 2012
17 11/13 Understanding Human Activities  
  • Gall, J., Fossati, A., and van Gool, L. (2011). Functional categorization of objects using real-time markerless motion capture. In CVPR.
  • Learning Human Activities and Object Affordances from RGB-D Videos, Hema S Koppula, Rudhir Gupta, Ashutosh Saxena. International Journal of Robotics Research (IJRR), 32(8):951-970, July 2013.
18 tba Understanding Human Activities   Joint estimation of object/scene affordances and human


  • V.Delatitre, D.Fouhey, I.Laptec, Sivic, A.Gupta, A.A.Efros "Scene semantics from long term observation of people" ECCV2012 
  • A.Gupta, S.Satkin, A.A.Efros, M.Hebert "From 3D scene geometry to Human workspace" CVPR2011 
  • People Watching: Human Actions as a Cue for Single View Geometry, David Fouhey, Vincent Delaitre, Abhinav Gupta, Alexei A. Efros, Ivan Laptev, and Josef Sivic European Conference on Computer Vision (ECCV), October, 2012.
Final Project Presentations
Final Project Presentations    
  11/25 no class - Thanksgiving Break    
  11/27 no class - Thanksgiving Break    

no class due to ICCV

  12/4 no class due to ICCV     
Final Project Report Due    

Personal tools