Cross-view activity recognition using Hankelets

By Binlong Li, Octavia I. Camps, Mario Sznaier. 2012

157 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Biomedical Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

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

In Artificial Intelligence

In Biomedical Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, methodology): “Since HOPC and STK-Distribution capture complimentary information, their combination significantly improves classification accuracy which is over 22% higher than the nearest competitor on both datasets.”
Cross-View Action Modeling, Learning, and Recognition (2014, background): “Hankelet [10] represents actions with the dynamics of short tracklets, and achieves cross-view action recognition by finding the Hankelets that are relative invariant to viewpoint changes.”
Glimpse Clouds: Human Activity Recognition from Unstructured Feature Points (2018, neutral): “Hankelets [31] V 45.”
Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons (2016, methodology): “Such constraints have been used to model actions; examples include but not limited to using Lie Alge-bra [27], positive deﬁnite matrices [3, 5], using a torus manifold [2], Hanklet representations [11], among several others.”
Learning a Deep Model for Human Action Recognition from Novel Viewpoints (2016, methodology): “of [13], [15], [23], [33], [52], [53], [54], [55] or from the original papers.”
Efficient Temporal Sequence Comparison and Classification Using Gram Matrix Embeddings on a Riemannian Manifold (2016, methodology): “where we follow [21] by using a Frobenius normalization.”
3D skeleton-based human action classification: A survey (2016, background): “Instead, a set of trajectories of the 3D joints is represented by an Hankelet [126], which embeds the observability matrix of the linear time invariant (LTI) system.”
Gesture Modeling by Hanklet-Based Hidden Markov Model (2014, methodology): “Hankel matrices have been successfully used in previous works on action recognition [41], tracking [43] and dynamic textures [42].”
A Review of Human Activity Recognition Methods (2015, methodology): “A new type of feature called the “hankelet” was presented by Li et al. (2012).”
3D Pose from Motion for Cross-View Action Recognition via Non-linear Circulant Temporal Encoding (2014, unknown): “Figure 4: We compare the performance of our approach (nCTE + data augmentation) with our baseline and Hankelets (Table 3 in [13]), the current state-of-the-art.”
Learning a non-linear knowledge transfer model for cross-view action recognition (2015, methodology): “|3 t1, 3u|2 t2, 3u|1 Mean Hankelets [18] 45.2 - - - DVV [19] 58.5 55.2 39.3 51.0 CVP [40] 60.6 55.8 39.5 52.0 AOG [33] 73.3 - - - nCTE [11] 68.6 68.3 52.1 63.0 Proposed NKTM 75.8 73.3 59.1 69.4 IXMAS [34] and N-UCLA [33] datasets by transferring knowledge using the same NKTM learned without supervis”
Cross-View Action Recognition via Transferable Dictionary Learning (2016, methodology): “proposed in [23] for recognizing actions across different viewpoints.”
The Way They Move: Tracking Multiple Targets with Similar Appearance (2013, background): “[23] introduced a different similarity score, based on the subspace angle between the column spaces of the Hankel matrices, that is efficient to compute and does not require estimating rank.”
Movement Pattern Histogram for Action Recognition and Retrieval (2014, methodology): “For this experiment we use 1NN classifier and a 6-fold cross validation procedure (identical cross-validation procedure as in [16,15,30]).”
Learning View-Invariant Sparse Representations for Cross-View Action Recognition (2013, background): “[14] proposed the Hankel matrix of a short tracklet which is a view-invariant feature to recognize actions across different viewpoints.”