Office of Naval Research: N000141010933

ONR MURI 2011-2012 Annual Meeting on

Knowledge Representation, Reasoning and Learning for Understanding Scenes and Events

[UCLA Sunset Canyon Recreation Center]

Session 1: Overview

8:15-8:25

Report on the Project theme and Progress in 2011-2012
Song-Chun Zhu

8:25-8:30

Comments from ONR
Behzad Kamghar-Parsi

Session 2: Object, Scene and Action Recognition

8:30-9:00

Large-scale learning of 3D object models
Deva Ramanan (UC Irvine)

9:00-9:30

Detecting and annotating 3D objects in scenes and panoramas
Antonio Torralba (MIT)

9:30-10:00

Rich and scalable models for scene understanding
Michael Stark and Daphne Koller (Stanford)

10:15-10:45

Pose and Actions in Behavior Analysis
Pietro Perona (Caltech)

10:45-11:15

Human Activity Recognition
Li Fei-Fei (Stanford)

11:15-11:45

The three R’s of Computer Vision: Recognition, Reconstruction and Reorganization
Jitendra Malik (UC Berkeley)

Session 3: Learning, Inference and Evaluation

1:00-1:30

Compositional Models
Alan Yuille (UCLA)

1:30-2:00

Learning Compositional Patterns in Sparse Representation
Yingnian Wu (UCLA)

2:00-2:30

A Restricted Turing Test
Stuart Geman (Brown)

Session 4: Knowledge Representation, Causality and Reasoning

2:45-3:15

Modeling intuitive physics and intuitive psychology for richer scene understanding
Josh Tenenbaum (MIT)

3:15-3:45

Generalizability in Causal Inference
Elias Barenboin and Judea Pearl (UCLA)

4:45-4:15

Knowledge Representation and inference by Spatial, Temporal, and Causal AOGs
Song-Chun Zhu (UCLA)

Session 5: Panel Discussion

4:30-5:00

Challenges and Collaboration Plan