Trajectory-Based Modeling of Human Actions with Motion Reference Points

By Yu-Gang Jiang, Qi Dai, Xiangyang Xue, Wei David Liu, Chong-Wah Ngo. 2012

224 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Control and Systems Engineering

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

In Computer Vision and Pattern Recognition

In Control and Systems Engineering

Citations in the corpus

Beyond Gaussian Pyramid: Multi-skip Feature Stacking for action recognition (2014, methodology): “Among them, the trajectory based approaches [18, 35, 38, 40, 8], especially the Dense Trajectory method proposed by Wang et al.”
Actions ~ Transformations (2015, background): “More recently, trajectory based approaches [51, 31, 29, 50, 18, 32, 25] have shown signiﬁcant improvement in action recognition.”
Better Exploiting Motion for Better Action Recognition (2013, methodology): “[9] used a one-against-one multi-class SVM, which might have resulted in inferior scores.”
Activity representation with motion hierarchies (2013, methodology): “Our method is outperformed on the Hollywood 2 dataset by [Vig et al, 2012], who use BOF with saliency masks, [Jiang et al, 2012a], who combine multiple features, and [Wang et al, 2013], who also combine multiple trajectory features and uses spatio-temporal pyramids.”
Video Visual Relation Detection (2017, background): “Both hand-crafted features [10, 37] and deep neural networks [35, 38] are developed to resolve this problem.”
High-level event recognition in unconstrained videos (2012, background): “Mostly recently, the work of [54] proposed to use local and global reference points to model the motion of dense trajectories, leading to a comprehensive representation that integrates trajectory appearance, location, and motion.”
Action Recognition with Actons (2013, methodology): “Moreover, recent work [27, 7, 6] demonstrate that leveraging the trajectory information [20, 18] leads to more discriminative feature representation and makes for recognition performance.”
AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks for Human Action Recognition in Videos (2016, methodology): “Later works improved upon this pipeline in several ways [23, 47] by using dense sampling for feature extractions [41], describing trajectories instead of 3D points [13, 39], and using better pooling and encoding methods [47, 25, 23].”
Discriminative Orderlet Mining for Real-Time Recognition of Human-Object Interaction (2014, unknown): “Recently, dense trajectory [15, 16] based algorithm achieved promising results on the challenging datasets like HMDB and Hollywood2.”
Self-Paced Learning with Diversity (2014, unknown): “2012 [26] 59.”
Human Action Recognition in Unconstrained Videos by Explicit Motion Modeling (2015, methodology): “This work extends upon a previous conference publication [43] by adding new experiments on a large dataset, more comparative analysis with alternative methods and baselines, and extra discussions throughout the paper.”
What do 15,000 object categories tell us about classifying and localizing actions? (2015, unknown): “Such techniques include sampling at interest points [7, 23], densely [40, 54] or along dense trajectories [9, 17, 28, 50].”
A Robust and Efficient Video Representation for Action Recognition (2015, background): “On Hollywood2, all presented results (Jain et al, 2013; Jiang et al, 2012; Mathe and Sminchisescu, 2012; Zhu et al, 2013) improve dense trajectories in different ways.”
A Cause and Effect Analysis of Motion Trajectories for Modeling Actions (2014, result): “Similar to the proposed work, few of the trajectory-based methods to perform action classification are presented in [1, 17, 29, 25, 18, 11].”
Going deeper into action recognition: A survey (2016, methodology): “The dominant motion is assumed to be caused by the camera and is compensated from original trajectories by subtraction in Jiang et al. (2012) and through afﬁne transformations in Jain et al. (2013).”