Discriminative subvolume search for efficient action detection

By Junsong Yuan, Zicheng Liu, Ying Wu. 2009 IEEE Conference on Computer Vision and Pattern Recognition, 2009

319 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/4c24176d8f3a25eb96aa874f663680970602b77b

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

Modeling the Temporal Extent of Actions (2010, background): “[13] propose a branch-and-bound algorithm which searches for a 3-D bounding box, akin to our temporal cropping, by maximizing mutual information of features and actions under a näıve Bayes assumption.”
Sparse reconstruction cost for abnormal event detection (2011, background): “Research in video surveillance has made great progresses in recent years, such as background model [22], object tracking [3], pedestrian detection [8], action recognition [27] and crowd counting [7].”
Spatio-Temporal Naive-Bayes Nearest-Neighbor (ST-NBNN) for Skeleton-Based Action Recognition (2017, methodology): “In [34], NBNN is re-designed as Naive-Bayes based Mutual Information Maximization (NBMIM) to solve action detection problem.”
Visual Semantic Role Labeling (2015, background): “Given annotations can be generated automatically, this will be a large dataset, but is inadequate for us as it does not have the visual grounding which is our interest here.”
Recognizing actions by shape-motion prototype trees (2009, result): “These results are also comparable in performance to recent results reported in [15,32].”
A Multi-stream Bi-directional Recurrent Neural Network for Fine-Grained Action Detection (2016, methodology): “Several algorithms have been proposed since then for detecting actions using space time interest points [33], multiple instance learning [9], or part-based models [25, 10].”
Measuring and reducing observational latency when recognizing actions (2011, background): “Evaluating data on video where the action is being performed continuously eliminates the ambiguity that occurs as the user transitions into different actions.”
Action recognition by dense trajectories (2011, background): “The HOF descriptors computed on YouTube videos are heavily polluted by camera motions, since many videos are collected by hand-held cameras.”
Fuzzy human motion analysis: A review (2014, unknown): “MSRAction 2009 [222] MSR3D Action 2010 [223] RA = 97.”
Exploring the Trade-off Between Accuracy and Observational Latency in Action Recognition (2012, background): “Some systems perform temporal segmentation (Cao et al. 2010; Yuan et al. 2009), but these systems also assume that the action has already been recorded in video.”
Abnormal event detection in crowded scenes using sparse representation (2013, background): “Related Work Much progresses in video surveillance have been achieved in recent years for some key areas, such as background model [29], object tracking [2], pedestrian detection [8], action recognition [36], crowd counting [6] and traffic monitoring [32].”
A database for fine grained activity detection of cooking activities (2012, background): “The MSR actions [43] pose a detection challenge limited to three classes.”
Spatio-temporal Object Detection Proposals (2014, unknown): “Efﬁcient search methods for spatio-temporal action localization have been proposed in the past, exploiting additivity structures of bag-of-word representations and linear classiﬁers, e.g . using branch-and-bound search or dynamic programming [38,48].”
Human Action Localization with Sparse Spatial Supervision (2017, background): “Initial attempts for temporal and spatio-temporal action localization are based on a sliding-window scheme and focus on improving the search complexity [23], [24], [25], [26].”