Completed
Patches: Interest Points
Class Central Classrooms beta
YouTube videos curated by Class Central.
Classroom Contents
Visual Analysis of Extremely Dense Crowded Scenes
Automatically move to the next video in the Classroom when playback concludes
- 1 Motivation
- 2 Presentation Layout
- 3 Key Ideas
- 4 Problems
- 5 Related Work: Counting by Detection
- 6 Related Work: Counting by Regression
- 7 Spatial Poisson Counting Process
- 8 Patches: Head Detections
- 9 Patches: Fourier Analysis
- 10 Patches: Interest Points
- 11 Patches: Fusion
- 12 Images: Multi-scale MRF
- 13 Results: Quantitative
- 14 Results: Per Patch Analysis
- 15 Results: Performance Analysis
- 16 Results: Analysis of 10th Group
- 17 Localization
- 18 Schematic Outline
- 19 Search results: Uniform Grid
- 20 Finding Representative Templates
- 21 Hypotheses Selection
- 22 Optimization
- 23 Bint Quadratic Programming
- 24 Background: Deformable Parts Model
- 25 Framework
- 26 Scale and Confidence Priors
- 27 Intermediate Results
- 28 Combination-of-Parts (COP) Detection
- 29 Global Occlusion Reasoning
- 30 Dataset: UCF-HDDC
- 31 Results: Qualitative
- 32 Results: Step-wise Improvement
- 33 Results: Density based Analysis
- 34 Results: Comparison
- 35 Results: Failure Cases
- 36 Chapter Summary
- 37 Queen Detection
- 38 Detection of Prominent Individuals
- 39 Modeling Crowd Behavior
- 40 Neighborhood Motion Concurrence
- 41 Tracking: Hierarchical Update
- 42 Experiments: Sequences
- 43 Quantitative Comparison
- 44 Component Contribution
- 45 Dissertation Conclusion
- 46 Future Work
