Spatio-temporal relationship match: Video structure comparison for recognition of complex human activities

By Michael S. Ryoo, Jake K. Aggarwal. IEEE International Conference on Computer Vision, 2009

624 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Biomedical Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/906e03d61d6a87b09cf7920d882f09fcedc8223d

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 Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Fast Exact Hyper-graph Matching with Dynamic Programming for Spatio-temporal Data (2014, neutral): “We use [36] N/A 5.1s N/A s/frame Savarese et al. [43] 86.8% N/A LOOCV Ryoo & Aggarwal [42] 93.8% N/A LOOCV Mikolajczyk & Uemura [38] 95.3% 5.5s to 8s LOOCV s/frame (-5s if SVM are not used) Baccouche et al. [2] 95.8 N/A LOOCV N/A Jiang et al. [22] 93.4% N/A LOOCV Our method 91.0% 0.0035s Split…”
Progressively Parsing Interactional Objects for Fine Grained Action Detection (2016, background): “Although general action recognition has a rich literature [19, 13, 23, 30, 35], fine-grained action analysis is a relatively new research direction.”
Every Moment Counts: Dense Detailed Labeling of Actions in Complex Videos (2015, background): “Interactions between humans or humans and objects (Ryoo and Aggarwal 2009; Oh et al. 2011) have been studied.”
Action Recognition by Hierarchical Mid-Level Action Elements (2015, methodology): “Moreover, we demonstrate the effectiveness of our model in real-world applications such as action recognition in large-scale untrimmed videos and action parsing.”
A review on vision techniques applied to Human Behaviour Analysis for Ambient-Assisted Living (2012, positive): “Based on similar ideas, Ryoo & Aggarwal [100], propose a kernel function to measure similarity between pairs of videos.”
Recognizing Micro-Actions and Reactions from Paired Egocentric Videos (2016, background): “Recognizing actions and reactions with a large amount of delay will require a structured prediction [28, 30, 37].”
Sequential Deep Learning for Human Action Recognition (2011, background): “g Hollywood-2 dataset [18], UCF sports action dataset [21], YouTube action dataset [16], UT-Interaction dataset [22] or LIRIS human activities dataset(2).”
Two-person interaction detection using body-pose features and multiple instance learning (2012, methodology): “Recognition of complex non-periodic activities, especially interactions between multiple persons, will be necessary for a number of applications (e.g. automatic detection of violent activities in smart surveillance systems).”
A database for fine grained activity detection of cooking activities (2012, background): “Surveillance datasets PETS 2007 [11] det 3 10 32,107 768x576 UT interaction [28] cls,det 6 120 6 VIRAT [23] det 23 17 1920x1080”
Action-Reaction: Forecasting the Dynamics of Human Interaction (2014, background): “Traditionally dual-agent interactions (e.g.., hugging, pushing) have been represented as a joint phenomenon, where observations from both people are used as features to recognize human interactions from video [22,2,21,8,26].”
High Five: Recognising human interactions in TV shows (2010, methodology): “Our experiments show that using structured learning improves the retrieval results compared to using the interaction classifiers independently.”
A Unified Framework for Multi-target Tracking and Collective Activity Recognition (2012, background): “[21] models the pairwise interactions between people, but the model is limited to local motion features.”
Real-time Action Recognition by Spatiotemporal Semantic and Structural Forests (2010, methodology): “The proposed method is tested on two public benchmarks, the KTH data set [23] and the UTinteraction data [20], a more challenging set [19].”
Human Action Recognition With Video Data: Research and Evaluation Challenges (2014, methodology): “Shape, Appearance [14], [17], [21], [22], [27], [30], [36], [37], [43], [48], [52], [53], [57]–[59], [65], [69], [91], [100], [124]–[128]”