Learning Motion Categories using both Semantic and Structural Information

By Shu-Fai Wong, Tae-Kyun Kim, Roberto Cipolla. 2007

239 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Human-Computer Interaction

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

In Human-Computer Interaction

Citations in the corpus

Learning and Refining of Privileged Information-Based RNNs for Action Recognition from Depth Sequences (2017, background): “Unlike most of the previous works that are based on a certain assumption about the structure of the depth maps or the availability of a robust skeleton tracker, our model automatically learns features from raw depth maps irrespective of any assumptions [58, 51] on the structure of video sequences .”
Unstructured human activity detection from RGBD images (2011, methodology): “We have also made the dataset and code available open-source at: http://pr.cs.cornell.edu/humanactivities”
Modeling Temporal Structure of Decomposable Motion Segments for Activity Classification (2010, background): “This representation can be combined with either discriminative [4,5] classifiers, semi-latent topic models [6] or unsupervised generative [7,8] models.”
Human Action Recognition and Prediction: A Survey (2018, methodology): “The method in [120] considers local visual words as parts, and models the structure information between parts.”
Spatiotemporal Deformable Part Models for Action Detection (2013, methodology): “[27] apply pLSA to capture the spatiotemporal relationship of visual words.”
H+O: Unified Egocentric Recognition of 3D Hand-Object Poses and Interactions (2019, background): “While action recognition is a longstanding problem in computer vision [3, 10, 18, 26, 31, 38, 69], first-person action recognition started to emerge as an active field only recently, largely due to the advent of consumer-level wearable sensors and large egocentric datasets [9, 20, 44, 52].”
H2O: Two Hands Manipulating Objects for First Person Interaction Recognition (2021, background): “Action recognition has received a lot of attention in the computer vision community [4, 14, 36, 42, 57, 84, 88].”
Recognizing 50 human action categories of web videos (2013, methodology): “Template based methods [1], modeling the dynamics of human motion using finite state models [6] or hidden Markov models [21], and Bag of Features models [10,4,11,22] (BOF) are a few well known approaches taken to solve action recognition.”
Cross-dataset action detection (2010, background): “In the past several years, much research work is based on modeling spatial-temporal interest points (STIPs) [17], [2], [6], [20], [11], [14].”
Slow Feature Analysis for Human Action Recognition (2012, methodology): “Savarese et al. [36] proposed the spatial-temporal correlograms to encode the long range temporal information into the local motion features.”
Spatio-temporal relationship match: Video structure comparison for recognition of complex human activities (2009, methodology): “Recently, spatio-temporal feature-based approaches have been considered by several researchers [13, 3, 16, 9].”
Human Action Recognition With Video Data: Research and Evaluation Challenges (2014, unknown): “[132], [133], [135], [156] × × × I × × √ ×”
Ongoing human action recognition with motion capture (2014, methodology): “[29] have proposed an extension to the pLSA space to model spatial relations [42].”
Understanding Video Events: A Survey of Methods for Automatic Interpretation of Semantic Occurrences in Video (2009, methodology): “A more recent group [33, 104, 105] of works make use of Bayesian Network models adapted from the”
Real-time Action Recognition by Spatiotemporal Semantic and Structural Forests (2010, methodology): “Since these methods [26, 28, 30] encode holistic structures with respect to a reference position e.”
Recognizing realistic actions from videos “in the wild” (2009, unknown): “The commonly used KTH dataset contains relatively complicated scenarios, and many methods employing this dataset have been reported [8, 9 ,10].”
Unsupervised learning of event AND-OR grammar and semantics from video (2011, background): “Event analysis has gone a long way from modeling transient or periodic action [10, 3], to longer events [13, 6, 8], achieving promising recognition performance.”
A Spatio-Temporal Descriptor Based on 3D-Gradients (2008, unknown): “[20] since their results are based on a non-standard setup.”