View Invariance for Human Action Recognition

By Vasu Parameswaran, Ramalingam Chellappa. 2005

237 citations recorded by the source corpus.

Subfields: 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 Computer Vision and Pattern Recognition

Citations in the corpus

Space-Time Representation of People Based on 3D Skeletal Data: A Review (2016, neutral): “[163] HD Models Conc Lowlv Deep X X X X Parameswaran and Chellappa [164] ISTs Conc Lowlv Hand X X X X”
Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, background): “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.”
HOPC: Histogram of Oriented Principal Components of 3D Pointclouds for Action Recognition (2014, methodology): “In RGB videos, in order to recognize actions across viewpoint changes, mostly view independent representations are proposed such as view invariant spatio-temporal features [2,20,22,23,25,33].”
Cross-View Action Modeling, Learning, and Recognition (2014, background): “Parameswaran [15] proposes viewinvariant canonical body poses and trajectories in 2D invariant space.”
Tensor Representations via Kernel Linearization for Action Recognition from 3D Skeletons (2016, methodology): “Action dynamics has been modeled using the motion of such 3D points in [14, 15], using joint orientations with respect to a reference axis [16] and even relative body-joint positions [17, 18].”
Cross-view action recognition via view knowledge transfer (2011, background): “[34] extracted view-invariant features by checking the planarity of the body joints.”
Learning a non-linear knowledge transfer model for cross-view action recognition (2015, background): “However, the discriminative power of these methods is limited by their inherent structure of view invariant features [38].”
A connectionist architecture for view-independent grip-aperture computation. (2008, background): “This aspect has received relatively scarce attention ( Parameswaran and Chellappa , 2006 ) so far , and the computational modelling of brain mechanisms which enable one to extract visual features of body parts in a view - independent fashion is needed to advance our understanding of action…”
Learning human actions via information maximization (2008, background): “al.[13] proposed an approach to exploit the 2D invariance in 3D to 2D projection, and model actions using view-invariant canonical body poses and trajectories in 2D invariance space.”
Human activities recognition using depth images (2013, unknown): “Copyright 2013 ACM 978-1-4503-2404-5/13/10 ...$15.00.”
View-invariant human action recognition via robust locally adaptive multi-view learning (2015, background): “Parameswaran and Chellappa (2006) calculated the geometric invariances of five key points on the same plane.”
3D Pose from Motion for Cross-View Action Recognition via Non-linear Circulant Temporal Encoding (2014, unknown): “Parameswaran and Chellappa [16] propose Invariance Space Trajectories, a geometrical invariant that uses 5 points that lie in the same plane.”
Cross-View Action Recognition from Temporal Self-similarities (2008, background): “[5] proposes a quasi view-invariant approach, requiring at least 5 body points lying on a 3D plane or that the limbs trace a planar area during the course of an action.”
Cross-View Action Recognition via a Continuous Virtual Path (2013, background): “Some of these approaches rely on geometric constraints [28], body joints detection and tracking [20, 21], and 3D models [14, 26].”
Action recognition in videos acquired by a moving camera using motion decomposition of Lagrangian particle trajectories (2011, background): “Poisson equation-based features [6], 3D Haar feature [7]), spatiotemporal interest points ([8, 11, 14]), motion history image (MHI) [15], and kinematic features [12].”
Action Recognition from Arbitrary Views using 3D Exemplars (2007, methodology): “To achieve similar comparisons, Parameswaran and Chellappa [14] use projective invariants of coplanar landmark points on a human body.”
A survey on activity recognition and behavior understanding in video surveillance (2013, background): “Parameswaran [132] developed a high-level representation and an efficient recognition algorithm for human action that was resistant to variation in viewpoints, and used projective invariants of landmark points on a human body.”