Real-time Action Recognition by Spatiotemporal Semantic and Structural Forests

By Tsz-Ho Yu, Tae-Kyun Kim, Roberto Cipolla. 2010

141 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

Citations in the corpus

Efﬁcient feature extraction, encoding and classiﬁcation for action recognition (unknown): “Our method demonstrates better classiﬁcation accuracy compared to [40] and improves the speed of [40] by one order of magnitude.”
A Discriminative Model with Multiple Temporal Scales for Action Prediction (2014, unknown): “Action Recognition: Human actions [17,26,15,3] have been popularly represented by a set of quantized local spatiotemporal features, known as bag-ofwords.”
Transition Forests: Learning Discriminative Temporal Transitions for Action Recognition and Detection (2016, methodology): “Previous approaches encode the temporal variable in the feature space by stacking multiple frames [10], handcrafting temporal features [34, 40] or creating codebooks [34].”
Real-Time Online Action Detection Forests Using Spatio-Temporal Contexts (2016, background): “…opportunities, however come with signiﬁcant challenges not encountered by the classical ofﬂine action recognition approaches [49, 42, 12, 40, 47, 44, 43, 39, 7, 15, 19, 50, 5, 10]: Realistic OAD applications require real-time processing that prevents the use of computationally expensive features.”
Dynamic Sampling Networks for Efficient Action Recognition in Videos (2020, methodology): “[47] utilized local appearance and structural information to achieve real-time action recognition.”
A survey on activity recognition and behavior understanding in video surveillance (2013, background): “In addition, it encourages localization of the multiple activities involving multiple persons and objects [156, 198, 222] from”
Online Real-Time Multiple Spatiotemporal Action Localisation and Prediction (2016, background): “[59] evaluate their real-time continuous action classification approach on the relatively simpler KTH [34] and UT-interaction [31] datasets.”
Human Action Recognition and Prediction: A Survey (2018, background): “Even though some of the early work such as [4], [6], [12], [49], [131] used action videos containing human interactions, they recognize actions in the same way as single-person action recognition.”
Efficient Feature Extraction, Encoding, and Classification for Action Recognition (2014, methodology): “As one contribution of this paper, we show that the use of sparse MPEG flow instead of the dense OF improves the speed of feature extraction by two orders of magnitude and implies only minor reduction in classification performance.”
Recognize Human Activities from Partially Observed Videos (2013, background): “Recently, research on activity recognition ha s been extended to more complex activity scenarios which involve multiple persons interacting with each other or objects [15, 20, 24].”
Human activity prediction: Early recognition of ongoing activities from streaming videos (2011, methodology): “Researchers are now graduating from recognizing simple human actions such as walking and running [16, 4, 2, 10, 6], and the field is gradually moving towards recognition of complex realistic human activities involving multiple persons and objects [12, 17, 18].”
Encouraging LSTMs to Anticipate Actions Very Early (2017, unknown): “Accuracy D-BoW [27] 85.0% I-BoW [27] 81.7% Cuboid SVM [26] 85.0% BP-SVM [19] 83.3% Cuboid/Bayesian [27] 71.7% DP-SVM [34] 14.6% Yu et al. [50] 83.3% Yuan et al. [51] 78.2% Waltisberg et al. [41] 88.0% Ours 90.0% Diff. to State-of-the-Art +2.0% Table 8.”