Recognizing human action in time-sequential images using hidden Markov model

By Junji Yamato, J. Ohya, K. Ishii. 1992

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, background): “Human action refers to a temporal sequence of primitive movements carried out by a person [55].”
Beyond Gaussian Pyramid: Multi-skip Feature Stacking for action recognition (2014, methodology): “For lab datasets where human poses or action templates can be reliably estimated, Dynamic Time Warping (DTP) [5], Hidden Markov Models (HMMs) [41] and Dynamic Bayesian Networks (DBNs) [23] are well studied methods for aligning actions that have speed variation.”
Every Moment Counts: Dense Detailed Labeling of Actions in Complex Videos (2015, methodology): “[38], using hidden Markov models for latent action state spaces.”
Activity recognition and abnormality detection with the switching hidden semi-Markov model (2005, methodology): “This has led to the widespread use of dynamic models such as the hidden Markov model (HMM) [23, 22].”
A survey on activity recognition and behavior understanding in video surveillance (2013, methodology): “[213] adopted a hidden Markov model (HMM) to represent and recognize the activities.”
Learning to Recognize Activities from the Wrong View Point (2008, background): “Primitives: Motions are typically seen as having a compositional character, with hidden Markov models used to recognize, among others: tennis strokes [ 45 ]; pushes [43]; and handwriting gestures [46].”
Dynamic Feature Selection for Online Action Recognition (2013, background): “A typical offline recognition system processes pre-segmented action sequences of a single action and incorporates information across the entire sequence [5-10].”
Recognition of Visual Activities and Interactions by Stochastic Parsing (2000, methodology): “One of the earlier attempts to use HMMs for recognition of activities is found in the work by Yamato et al. [ 40 ], where discrete HMMs are used to recognize six tennis strokes, performed by three subjects.”
Video-based event recognition: activity representation and probabilistic recognition methods (2004, methodology): “In one of the earlier attempts [12], discrete HMMs are used as the representations of tennis strokes.”
Sequence of the Most Informative Joints (SMIJ): A new representation for human skeletal action recognition (2012, background): “[11] and a review of several others in [12].”
A survey of temporal data mining (2006, background): “…represent, e. g., one of a fixed set of messages like waving hello, goodbye, and so on (Darrell & Pentland 1993), or they could be the different strokes in a tennis video (Yamato et al 1992), or in other cases, they could belong to the dictionary of some sign language (Starner & Pentland 1995) etc.”
A survey of vision-based methods for action representation, segmentation and recognition (2011, methodology): “Figure 4: Global posture representations: (a) Silhouettes of tennis strokes [144]; (b) silhouettes pixels accumulated in regular grid [130]; (c) spline contours [98]; (d) optical flow magnitude accumulated in regular grid [94]; (e) optical flow split into directional components, then blurred [25].”
A real time system for dynamic hand gesture recognition with a depth sensor (2012, background): “So far, most of the research work has been focused on recognition from images and videos due to the popularity of digital cameras.”
Grounding the Lexical Semantics of Verbs in Visual Perception using Force Dynamics and Event Logic (1999, background): “ Yamoto et al. (1992) , Starner (1995), Siskind an d Morris (1996), Siskind (1996), Brand (1996, 1997a), Brand, Oliver, and Pentland (1997), and Bobick and Ivanov (1998) present systems that recognize event occurrences from real video using motion profile but not force dynamics.”
Human Action Recognition by Semilatent Topic Models (2009, unknown): “" We note that there has been previous work (e.g., Yamato et al. [9], Bobick and Wilson [10], Xiang and Gong [11]) that tries to model the full dynamics of videos using sophisticated probabilistic models (e.g., hidden Markov models and dynamic Bayesian networks).”
Structured Label Inference for Visual Understanding (2018, background): “Early attempts of modelling time-sequential images in computer vision rely on hidden Markov models for obtaining representations in order to recognize action classes considering evolution of actions or even multiple view-points [56, 33, 43].”