Motion history images for action recognition and understanding /
"Human action analysis and recognition is a relatively mature field, yet one which is often not well understood by students and researchers. The large number of possible variations in human motion and appearance, camera viewpoint, and environment, present considerable challenges. Some important...
| Main Author: | |
|---|---|
| Corporate Author: | |
| Format: | eBook |
| Language: | English |
| Published: |
London ; New York :
Springer,
[2013]
|
| Series: | SpringerBriefs in computer science.
|
| Subjects: | |
| Online Access: | Connect to the full text of this electronic book |
Table of Contents:
- 1. Introduction
- Action/Activity: Nomenclature
- Atomic Actions
- Action
- Activity
- Various Dimensions of Action Recognition
- Applications
- Action Recognition is Difficult: Why?
- Some Assumptions on Action Recognition
- Action Recognition: Some Basic Steps
- Motion History Image
- 2. Action Representation
- Action Recognition
- Approaches on Bag-of-Features
- XYT: Space-Time Volume
- Spatio-Temporal Silhouettes
- Interest-Point Detectors
- Local Discriminative Approaches
- Large-Scale Features-Based Recognition
- Local Patches-Based Recognition
- Mixed Approach for Recognition
- View-Invariant Approaches
- 3. Motion History Image
- Motion History Image and its Importance
- Applications of the MHI Method/Representation
- Motion History Image: A Tutorial
- Construction of the Motion Energy Image (MEI)
- Construction of the Motion History Image (MHI)
- Importance of the MHI and the MEI
- Limitations of the MHI Method
- MHI and Its Variants: 2D/3D
- View-Based MHI and Its Variants
- View-Invariant MHI Representations
- 4 Action Datasets and MHI
- Various Datasets
- Datasets Employed in MHI
- KTH Dataset
- Weizmann Dataset
- IXMAS Dataset
- CASIA Gait Database
- Virtual Human Action Silhouette (ViHASi) Dataset
- CMU MoBo Dataset
- USF HumanID Dataset
- Marcel's Dynamic Hand Poster and Gesture Dataset
- TRECVID Dataset.