2013 IEEE Conference on Computer Vision and Pattern Recognition Sampling Strategies for Real-time Action Recognition

Citations in the corpus

Beyond Gaussian Pyramid: Multi-skip Feature Stacking for action recognition (2014, neutral): “[31] 83.”
Action Classification with Locality-Constrained Linear Coding (2014, background): “Most of the previous approaches to action recognition have focused on using traditional RGB cameras [1]–[3].”
Efﬁcient feature extraction, encoding and classiﬁcation for action recognition (unknown): “We experimentally compare our method to [34] and show one order of magnitude improvement in speed while also demontrating improved accuracy.”
Efficient Feature Extraction, Encoding, and Classification for Action Recognition (2014, methodology): “We evaluate using kd-forest approximate nearest neighbor search [25] and analyze the associated trade-off between the speed and recognition accuracy.”
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 Cause and Effect Analysis of Motion Trajectories for Modeling Actions (2014, background): “Changing the delay parameter for smaller model orders did not affect the performance while for larger order values the performance degraded with increase in delay due to the decrease in number of equations for computing prediction matrices.”
Human Action Recognition in Unconstrained Videos by Explicit Motion Modeling (2015, background): “A few others focused on the selection or sampling of more effective local volumes [17], [18] or higher-level attribute representations [19], [20].”
Action Recognition with Actons (2013, background): “, which are computed on the local 3D patches extracted by spatial-temporal interesting point (STIP) detectors [11] or dense sampling strategies [27, 24].”
A Critical Review of Action Recognition Benchmarks (2013, methodology): “Curiously, despite being saturated and despite their simplistic settings, these benchmarks are still frequently used today, with recent examples including [13] on the Weizmann set and [34] on KTH.”