Joint Angles Similarities and HOG2 for Action Recognition

By Eshed Ohn-Bar, Mohan Manubhai Trivedi. 2013

326 citations recorded by the source corpus.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition, Human-Computer Interaction

How researchers see this paper

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

In Biomedical Engineering

1. Jun Liu
2. G. Wang
3. Hossein Rahmani
4. Wanqing Li
5. Pichao Wang
6. Amir Shahroudy
7. Alex ChiChung Kot
8. Qiuhong Ke

In Computer Vision and Pattern Recognition

1. Jun Liu
2. G. Wang
3. Hossein Rahmani
4. Wanqing Li
5. Pichao Wang
6. Amir Shahroudy
7. Alex ChiChung Kot
8. Qiuhong Ke

In Human-Computer Interaction

1. Jun Liu
2. Pichao Wang
3. Wanqing Li
4. Hossein Rahmani
5. Amir Shahroudy
6. G. Wang
7. Philip O. Ogunbona
8. Zhimin Gao

Citations in the corpus

• NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis (2016, methodology): “Ohn-Bar and Trivedi [24] combined second order joint-angle similarity representations of skeletons with a modified two step HOG feature on spatio-temporal depth maps to build global representation of each video sample and utilized a linear SVM to classify the actions.”
• NTU RGB+D 120: A Large-Scale Benchmark for 3D Human Activity Understanding (2019, methodology): “Ohn-Bar and Trivedi [55] combined second order joint-angle similarity representations of skeletons with a modified two step HOG feature on spatio-temporal depth maps to build global representation of each video sample, and utilized a linear SVM to classify the actions.”
• Deep Multimodal Feature Analysis for Action Recognition in RGB+D Videos (2016, neutral): “HOG [87] 32.”
• 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.”
• Multimodal Multipart Learning for Action Recognition in Depth Videos (2015, methodology): “ [5]) Accuracy Action Graph on Bag of 3D Points [5] 74.7% Histogram of 3D Joints [7] 79.0% EigenJoints [6] 83.3% Random Occupancy Patterns [8] 86.5% Depth HOG [38] 91.6% Lie Group [39] 92.5% JAS+HOG2 [40] 94.8% DL-GSGC+TPM [41] 96.7% Proposed MMMP 98.2% into 88.13%. Finally by adding the multipart l”
• Modeling Temporal Dynamics and Spatial Configurations of Actions Using Two-Stream Recurrent Neural Networks (2017, background): “While body part based approaches regard the human skeleton as a connected set of segments, and then focus on individual or connected pairs of body parts [50] and joint angles [33].”
• Chained Multi-stream Networks Exploiting Pose, Motion, and Appearance for Action Classification and Detection (2017, neutral): “, compares favorably Methods Cross Subject % Deep LSTM [32] 60.7% P-LSTM [32] 62.93% HOGˆ2 [25] 32.2% FTP DS [10] 60.23% ST-LSTM [21] 69.2% Ours (Pose) 67.8% Ours (RGB+OF+Pose - Baseline) 76.9% Ours (RGB+OF+Pose - Chained) 80.8 % to the existing methods.”
• Unsupervised Learning of Long-Term Motion Dynamics for Videos (2017, methodology): “We split the 40 subjects into training and testing groups Methods Depth HOG [17] 32.24 Super Normal Vector [42] 31.82 HON4D [18] 30.56 Lie Group [32] 50.08 Skeletal Quads [3] 38.62 FTP Dynamic Skeletons [4] 60.23 HBRNN-L [2] 59.07 2 Layer P-LSTM [25] 62.93 Shuffle and Learn [15] 47.5 Our method…”
• Spatio-Temporal LSTM with Trust Gates for 3D Human Action Recognition (2016, background): “. arXiv:1607.07043v1 [cs.CV] 24 Jul 2016 2 Jun Liu, Amir Shahroudy, Dong Xu, and Gang Wang proposed by [8]. An angular skeletal representation over the tree-structured set of joints was introduced in [9], which calculated the similarity of these features over temporal dimension to build the global r”
• Action Recognition Based on Joint Trajectory Maps Using Convolutional Neural Networks (2016, methodology): “lts compared with both hand-crafted features and deep learning methods. The work [34] focused only on single person action and could not model multi-person interactions well. Dynamic Skeletons method [38] performed better than some RNN-based methods verifying the weakness of the RNNs [43], [46], whi”
• Combined shape analysis of human poses and motion units for action segmentation and recognition (2015, unknown): “Likewise, in [11] skeleton features based on joint angles and depth features computed with histogram of oriented gradients are combined together.”
• Vision-based human activity recognition: a survey (2020, methodology): “Joint angles, which assume the human body as a system of rigid connected segments and the movement as an evolution of their spatial configuration [7, 29, 128].”
• Fusing Geometric Features for Skeleton-Based Action Recognition Using Multilayer LSTM Networks (2018, unknown): “12% − − − − HOG2 [24] − 32.”
• Evaluating spatiotemporal interest point features for depth-based action recognition (2014, background): “JAS (cosine) + MaxMin + HOG2 [36] 94.”
• Informative joints based human action recognition using skeleton contexts (2015, methodology): “In this paper, we introduce a robust informative joints based human action recognition method.”