Recognizing realistic actions from videos “in the wild”

By Jingen Liu, Jiebo Luo, Mubarak Shah. 2009

1,201 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Hollywood in Homes: Crowdsourcing Data Collection for Activity Understanding (2016, methodology): “In order to scale up the learning and the complexity of the data, recent approaches have instead tried collecting video datasets by downloading videos from Internet.”
Improving Interpretability of Deep Neural Networks with Semantic Information (2017, methodology): “We use the UCF11 dataset [22], a YouTube action dataset consisting of 1600 videos and 11 actions, including bas-ketball shooting, biking, diving, golf swinging, horse riding, soccer juggling, swinging, tennis swinging, trampoline jumping, volleyball spiking, and walking.”
Learning human activities and object affordances from RGB-D videos (2012, methodology): “There has been a lot of work on human activity detection from images (Yang et al., 2010; Yao et al., 2011) and from videos (Laptev et al., 2008; Liu et al., 2009; Hoai et al., 2011; Shi et al., 2011; Matikainen et al., 2012; Pirsiavash and Ramanan, 2012; Rohrbach et al., 2012; Sadanand and Corso,…”
Video In Sentences Out (2012, methodology): “, 2008] and the Youtube dataset [Liu et al., 2009], often make action-class distinctions that are irrelevant to the choice of verb, e.”
Action Recognition with Stacked Fisher Vectors (2014, methodology): “Following [20], we use Leave-One-Group-Out cross-validation and report the average accuracy over all classes.”
Detecting People Looking at Each Other in Videos (2013, neutral): “…(Ferrari et al. 2009; Andriluka et al. 2009; Bourdev et al. 2010; Sapp et al. 2010;Yang et al. 2012), human action recognition (Laptev et al. 2008; Liu et al. 2009; Marín-Jiménez and Pérez de la Blanca 2012; Raptis et al. 2012; Sadanand and Corso 2012), and social (Fathi et al. 2012) and speciﬁc…”
Web-Based Classifiers for Human Action Recognition (2012, methodology): “To evaluate this case, we use the UCFYouTube dataset [12].”
Does Human Action Recognition Benefit from Pose Estimation? (2011, methodology): “As recent trends in action recognition have shifted towards analysis in natural and unconstrained videos, such as sequences from feature films [17], broadcast sports [21] and Youtube [19], efforts have moved from high-level modelling of the human body to directly classifying actions with abstract an”
A large-scale benchmark dataset for event recognition in surveillance video (2011, background): “Datasets have been developed for movies [11] and sports [12], but, these actions and scene conditions do not apply effectively to surveillance videos.”
Depth Induced Feature Representation for 4D Human Activity Recognition (2014, background): “Much effort has been made in human activity understanding since human activities play important roles on smart healthcare and wellbeing [9], human-computer interfaces [30], video surveillance [19], and content-based video indexing.”
Fine-Grained Activity Recognition with Holistic and Pose Based Features (2014, result): “This is in contrast to existing activity recognition datasets [13,26,17,16,20] that typically include ad-hoc selections of activity classes.”
Video-Based Human Behavior Understanding: A Survey (2013, methodology): “The authors showed that motion boundaries encoded along the trajectories can outperform state-of-the-art trajectory descriptors, on the KTH [75], Youtube [96], Hollywood2 [97], and UCF Sport datasets [98].”
Discriminative subvolume search for efficient action detection (2009, methodology): “Contemporary methods for action categorization use local spatio-temporal features to characterize the video and perform classification over the set of local features [13] [12] [17] [19] To improve the classification performance, both shape and motion information are applied for action categorization”
Fuzzy human motion analysis: A review (2014, neutral): “UCF-11 Youtube 2009 [218] – – RA1⁄489.”
Action recognition by dense trajectories (2011, methodology): “We, then, briefly present the bag-offeatures model used for evaluating our dense trajectory features as well as the KTL tracking baseline.”