Mining 3D Key-Pose-Motifs for Action Recognition

By Chun-yu Wang, Yizhou Wang, Alan L. Yuille. 2016

80 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Control and Systems Engineering, 3D Vision & Pose

How researchers see this paper

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

In Computer Vision and Pattern Recognition

1. Luc Van Gool
2. Jiwen Lu
3. Jie Zhou
4. Junsong Yuan
5. Dacheng Tao
6. Tae-Kyun Kim
7. Xiaogang Wang
8. G. Wang

In Control and Systems Engineering

1. Guillermo Garcia-Hernando
2. Tae-Kyun Kim
3. Seungryul Baek
4. Hassen Drira
5. Amor Ben Tanfous
6. Zhiwu Huang
7. Luc Van Gool
8. Jiwen Lu

In 3D Vision & Pose

1. Tae-Kyun Kim
2. Luc Van Gool
3. Zhiyuan Shi
4. Guillermo Garcia-Hernando
5. Jiwen Lu
6. Chaoqun Weng
7. Boulbaba Ben Amor
8. Seungryul Baek

Citations in the corpus

• Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates (2017, neutral): “0% Key-Pose-Motifs Mining[87] Geometric 93.”
• Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, methodology): “Results on Action3D : We follow a similar evaluation protocol from [45, 46].”
• Deep Learning on Lie Groups for Skeleton-Based Action Recognition (2016, background): “Due to the development of depth sensors, 3D human activity analysis [26, 44, 22, 42, 40, 3, 41, 36, 43, 25, 34, 17] has attracted more interest than ever before.”
• First-Person Hand Action Benchmark with RGB-D Videos and 3D Hand Pose Annotations (2017, methodology): “We compare our dataset with other egocentric datasets in Section 3.5.”
• Coding Kendall's Shape Trajectories for 3D Action Recognition (2018, methodology): “For Florence3D-Action and UTKinect-Action datasets, we followed an additional setting for each: Leave-one-actor-out (LOAO) [32, 37] and Leave-one-sequence-out (LOSO) [42], respectively.”
• Recognizing Human Actions as the Evolution of Pose Estimation Maps (2018, background): “Hand-crafted features [42, 47, 13] were designed for describing evolution of 3D poses.”
• Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition (2018, background): “Thus, skeleton-based action recognition has attracted more and more attention in recent years [2–8].”
• Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action Recognition (2017, methodology): “While in [27], sequential pattern mining method is used to model temporal structures of a set of key poses.”
• Transition Forests: Learning Discriminative Temporal Transitions for Action Recognition and Detection (2016, methodology): “Following previous work [28, 30], we adopt a leave-one-subject-out protocol, e.”