Object, Scene and Actions: Combining Multiple Features for Human Action Recognition

By Nazli Ikizler-Cinbis, Stan Sclaroff. European Conference on Computer Vision, 2010

318 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/c7de6028a3b6c07a5544b48e132862923d9c01bd

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

Action recognition by dense trajectories (2011, background): “Trajectory descriptors, which only describe the motion of the trajectories, give surprisingly good results by themselves, e.g. 90.2% on KTH and 47.7% on Hollywood2 for dense trajectories.”
Visual event recognition in videos by learning from web data (2010, background): “details), few of them [5], [16], [17], [28], [30] focused on event recognition in the highly unconstrained consumer video domain.”
Recognizing human actions by attributes (2011, background): “Some approaches have attempted to integrate contextual information, such as spacetime neighborhood features[13], object-scene-action mixture models[12], and spatial and temporal relations[22].”
Prediction of Human Activity by Discovering Temporal Sequence Patterns (2014, unknown): “Index Terms—Activity prediction, causality, context-cue, predictability Ç”
Web-Based Classifiers for Human Action Recognition (2012, methodology): “In this dataset, we first extract the human tracks by using a “tracking-by-detection” scheme, following [53].”
What do 15,000 object categories tell us about classifying and localizing actions? (2015, unknown): “The pose of the person allows modeling human-object interactions [13, 35, 57].”
Fast action proposals for human action detection and search (2015, background): “Our work is also relevant to previous work [9, 10] that utilize human detector for human tracking and action localization.”
Video In Sentences Out (2012, background): “ng diving by correlation with blue since it ‘happens in a pool’ (Liu et al., 2009, p. 2002) or confusing basketball and volleyball ‘because most of the time the [:::] sports use very similar courts’ (Ikizler-Cinibis and Sclaroff, 2010, p. 506)). 2 The mind’s eye corpus Many existing video corpora us”
Action Recognition with Image Based CNN Features (2015, background): “This is beneﬁcial for describing complex human actions [18, 17, 9].”
“Important stuff, everywhere!” Activity recognition with salient proto-objects as context (2014, background): “Ikizler-Cinbis and Sclaroff [13] assume large moving regions as candidate objects.”
Action Recognition with Stacked Fisher Vectors (2014, unknown): “[5] 75.”
Learning to recognize objects in egocentric activities (2011, methodology): “More recently, Multiple Instance Learning (MIL) has shown great promise as a method for weakly supervised learning in computer vision communities [1, 9, 5, 23, 6].”
Recognizing 50 human action categories of web videos (2013, methodology): “[7] used multiple features from the scene, object, and person, and combined them using a Multiple MIL (multiple instance learning) approach.”
An efficient part-based approach to action recognition from RGB-D video with BoW-pyramid representation (2013, background): “Some existing works defined context as scenes or objects associated with actions [22–24].”