Real-Time RGB-D Activity Prediction by Soft Regression

By Jianfang Hu, Weishi Zheng, Lianyang Ma, G. Wang, J. Lai. 2016

93 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

1. Jun Liu
2. G. Wang
3. Hossein Rahmani
4. Qiuhong Ke
5. Amir Shahroudy
6. Alex ChiChung Kot
7. Ling-yu Duan
8. Mohammed Bennamoun

Citations in the corpus

• Skeleton-Based Online Action Prediction Using Scale Selection Network (2019, background): “Most of the existing works [1], [3], [4] focus on action prediction in wellsegmented videos, for which each video contains exactly one action instance.”
• Skeleton-Based Human Action Recognition With Global Context-Aware Attention LSTM Networks (2017, methodology): “We can observe that our GCA-LSTM network surpasses the state-ofthe-art skeleton-based methods in [29], [71], and [80], which”
• Skeleton-Based Action Recognition Using Spatio-Temporal LSTM Network with Trust Gates (2017, neutral): “LAFF (SKL) [90] Geometric 54.”
• View Adaptive Recurrent Neural Networks for High Performance Human Action Recognition from Skeleton Data (2017, neutral): “Methods setting-1 setting-2 LAFF [14] – 54.2 Dynamic Skeletons [13] 75.5 76.9 VA-LSTM 76.9 77.5 and s trans are on as in Fig.”
• View Adaptive Neural Networks for High Performance Skeleton-Based Human Action Recognition (2018, unknown): “Method setting-1 setting-2 LAFF [13] 54.”
• Skeleton-Based Action Recognition with Spatial Reasoning and Temporal Stack Learning (2018, result): “rms the state-of-the-art approach [32] by about 3.8% and 4.4% for setting-1 and setting-2, respectively. Table 2.The comparison results on SYSU dataset in accuracy (%) Methods Setting-1 Setting-2 LAFF[9](2016) - 54.2 DynamicSkeletons [8](2015) 75.5 76.9 VA-LSTM[32](2017) 76.9 77.5 SR-TSL(Ours) 80.7 ”
• Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition (2018, unknown): “LAFF(SKL) [53] 54.”
• Jointly Learning Heterogeneous Features for RGB-D Activity Recognition (2015, background): “Bearing on these merits, it is believed that the introduced depth information can greatly light up the research of human activity analysis [12], [24], [36].”
• Part-Activated Deep Reinforcement Learning for Action Prediction (2018, background): “Human activity analysis has aroused much attention in computer vision due to its broad prospects of applications [8, 17, 19, 29, 36].”
• Global Regularizer and Temporal-Aware Cross-Entropy for Skeleton-Based Early Action Recognition (2018, methodology): “We compared with [34] using their reported results based on skeleton.”
• SSNet: Scale Selection Network for Online 3D Action Prediction (2018, background): “[23] explored to incorporate 3D skeleton information for real-time action prediction in wellsegmented sequences, i.”
• Deep Bilinear Learning for RGB-D Action Recognition (2018, methodology): “We evaluated our methods on two public benchmark 3D action datasets: NTU RGB+D Dataset [22] and SYSU 3D HOI dataset [14], with two different evaluation protocols employed in each set.”
• Else-Net: Elastic Semantic Network for Continual Action Recognition from Skeleton Data (2021, background): “, human-robot interaction, selfdriving vehicles, and security surveillance [34, 6].”