Human Action Recognition by Shape Analysis of Motion Trajectories on Riemannian Manifold

By Maxime Devanne, Hazem Wannous, Stefano Berretti, Pietro Pala, Mohamed Daoudi, A. del Bimbo. 2014

Subfields: Computer Vision and Pattern Recognition, Representation & Self-Supervised Learning

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 Representation & Self-Supervised Learning

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, methodology): “[185] introduced a human representation by comparing the similarity”
When Kernel Methods Meet Feature Learning: Log-Covariance Network for Action Recognition From Skeletal Data (2017, methodology): “Among others, we can mention histogram based representations to perform temporal pooling [42, 41, 6], extraction of local spatio-temporal features from the data [4, 30], also applying bag-of-words or Fisher vector approaches to aggregate the raw joint representation in a unique action descriptor [6,”
Spatio-Temporal LSTM with Trust Gates for 3D Human Action Recognition (2016, methodology): “Accuracy Histogram of 3D Joints [5] 90.9% Grassmann Manifold [65] 88.5% Riemannian Manifold [66] 91.5% ST-LSTM (Joint Chain) 91.0% ST-LSTM (Tree) 92.4% ST-LSTM (Tree) + Trust Gate 97.0% Finally, we evaluate the classiﬁcation performance on early stopping conditions by feeding the ﬁrst p (0 p 1)…”
Efficient Temporal Sequence Comparison and Classification Using Gram Matrix Embeddings on a Riemannian Manifold (2016, unknown): “[10] 87.”
Mining 3D Key-Pose-Motifs for Action Recognition (2016, unknown): “Methods Acc (%) Year Maxime [5] 91.”
Learning features combination for human action recognition from skeleton sequences (2017, methodology): “…Florence 3D Actions [24] 88.2% — — — [27] — — 90.92% ± 1.74% — [12] 96.7% — — — [14] 88.36% 82.15% ± 4.18% — — [18] — — — 82.0% [21] 88.89% — — — [4] 92.1% 87.28% ± 2.41% 91.5% 87.04% [11] 95.62% — — — [23] 89.48% — 97.08% 90.88% [28] 93.09% — — — [20] 91.21% — 88.5% — [22] 92.03% — — — [8] 92 •…”
Recognizing Actions in 3D Using Action-Snippets and Activated Simplices (2016, unknown): “(Devanne et al. 2015) 91.”
Spatio-Temporal Graph Convolution for Skeleton Based Action Recognition (2018, methodology): “…system, we modify the origin of the coor-Table Method Accuracy Multi-part Bag-of-Poses (Seidenari et al. 2013) 82.00% Riemannian Manifold (Devanne et al. 2015) 87.04% Lie Group (Vemulapalli, Arrate, and Chellappa 2014) 90.88% Graph-Based (Wang et al. 2016a) 91.63% MIMTL (Yang et al.…”
Action Recognition Based on Joint Trajectory Maps with Convolutional Neural Networks (2016, background): “Joint representation captures the correlation of the body joints by extracting spatial descriptor [30], [56]– [59], geometric descriptor [33], [34], [39]–[41], [60] or key poses [35], [61]–[63].”
3D skeleton-based human action classification: A survey (2016, background): “A few other works [135,114,65] have adopted this other validation protocol.”