POCV 2014

The Ninth IEEE Computer Society Workshop on Perceptual Organization in Computer Vision

June 23, 2014 Columbus, Ohio, USA In Conjunction with IEEE CVPR 2014
7:30 - 8:30
Breakfast
8:30 - 10:10
Session 1: Segmentation & Video
8:30
Welcome
8:30 - 8:50
Moving vs. Static Objects in Video Segmentation
Thomas Brox (University of Freiburg)
[Slides]
8:50 - 9:10
Composite Statistical Learning and Inference in Semantic and Video Segmentation
Fuxin Li (Georgia Institute of Technology)
[Slides]
9:10 - 9:30
Interactive Learning for Point-Cloud Motion Segmentation
Tal Hassner (Open University of Israel)
[Slides]
9:30 - 9:50
Learning from YouTube Videos
Rahul Sukthankar (Google)
9:50 - 10:05
Spotlight Presentations
Unsupervised Learning Using the Tensor Voting Graph
Shay Deutsch and Gerard Medioni
Intersecting Manifolds: Detection, Segmentation, and Labeling
Shay Deutsch and Gerard Medioni
Robust Symmetry Detection in Natural Images
Ching L. Teo, Hyoungjune Yi, and Cornelia Fermuller
[Slides]
10:05 - 10:30
Coffee Break
10:30 - 12:00
Session 2: Representation
10:30 - 10:50
From Edges to Objects
Piotr Dollar (Microsoft Research)
[Slides]
10:50 - 11:30
Perceiving Crowds: Bypassing the Bottleneck of Conscious Vision
David Whitney (University of California, Berkeley)
11:30 - 12:00
Hierarchy, Reasoning, and Representation Learning
Yann LeCun (New York University)
12:00 - 13:30
Lunch Buffet
13:30 - 15:00
Session 3: Activity
13:30 - 13:50
Live Counting of Repetitive Actions via a Convolutional Deep Neural Network Trained on Unrealistic Synthetic Data
Lior Wolf (Tel Aviv University)
13:50 - 14:10
Weakly-Supervised Learning of Actions
Ivan Laptev (INRIA)
[Slides]
14:10 - 14:30
Recognizing Human Activities: Interactions, Groups, and Context
Greg Mori (Simon Fraser University)
[Slides]
14:30 - 14:50
Video Scene Segmentation and Recognition by Location-Independent Activity Classes
Anthony Hoogs (Kitware)
14:50 - 15:00
Spotlight Presentations
Contour Motion Estimation for Address-Event Data with DVS
Francisco Barranco, Cornelia Fermuller, and Yiannis Aloimonos
Shadow-free Segmentation in Still Images Using Local Density Measure
Aleksandrs Ecins, Cornelia Fermuller, and Yiannis Aloimonos
[Slides]
15:00 - 15:20
Coffee Break
15:20 - 17:00
Session 4: Motion, Shape, & Parts
15:20 - 15:40
From Shallow Hollywood Motions to Deep Academic Gestures
Christoph Bregler (New York University)
15:40 - 16:20
Perceptual Organization of Motion: Event Recognition and the Perception of 3D Structure from Motion
James Todd (Ohio State University)
16:20 - 16:40
Do Mid-Level Parts Still Matter in the Age of CNNs?
Lubomir Bourdev (Facebook)
16:40 - 17:00
Data-Driven Perceptual Interpretation of Shape
William Freeman (Massachusetts Institute of Technology)