Dynamic Feature Selection for Online Action Recognition

By Victoria Bloom, Vasileios Argyriou, Dimitrios Makris. 2013

37 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Wanqing Li
2. Pichao Wang
3. Tae-Kyun Kim
4. Philip O. Ogunbona
5. Seungryul Baek
6. Kwang In Kim
7. Dimitrios Makris
8. Vasileios Argyriou

In Computer Vision and Pattern Recognition

1. Wanqing Li
2. Pichao Wang
3. Philip O. Ogunbona
4. Jun Liu
5. Tae-Kyun Kim
6. G. Wang
7. Chang-Fu Tang
8. Seungryul Baek

Citations in the corpus

• RGB-D-based action recognition datasets: A survey (2016, background): “Gaming 3D dataset (G3D) [16, 17]( http://dipersec.king.ac.uk/G3D/ ) captured by Kingston University in 2012 focuses on real-time action recognition in gaming scenario.”
• Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, background): “Many other existing human representations are also based on low-level skeleton-based features [117], [118], [119], [123], [134], [139], [140], [141], [154], [157], [158], [162], [200] without modeling the hierarchy of the data.”
• Real-Time Online Action Detection Forests Using Spatio-Temporal Contexts (2016, methodology): “OAD is a challenging problem [11, 25] and even though skeleton-based approaches [1, 52, 29, 35, 26, 25] are promising, they are bounded by the time constraint that prevents combining multiple feature modalities or considering more temporal relationships.”
• Transition Forests: Learning Discriminative Temporal Transitions for Action Recognition and Detection (2016, background): “Some early approaches [2, 10] include using short sequences of frames or short motion information (MP) [35] to vote if an action is being performed.”
• Real-Time Multi-scale Action Detection from 3D Skeleton Data (2015, methodology): “Results in [2] were reported for the “Fighting” action sequences only, however, we list our results for all seven action sequences in the supplementary materials.”
• G3Di: A Gaming Interaction Dataset with a Real Time Detection and Evaluation Framework (2014, methodology): “A novel realistic and challenging human interaction dataset, G3Di for real time multiplayer gaming is introduced.”
• Clustered Spatio-temporal Manifolds for Online Action Recognition (2014, methodology): “The higher accuracy of our method may be attributed to the improved execution rate invariance gained by matching template fragments with DTW instead of a fixed size feature window as used by both [14] and [15].”