Cross-view action recognition via view knowledge transfer

By Jingen Liu, Mubarak Shah, Benjamin Kuipers, Silvio Savarese. 2011

289 citations recorded by the source corpus.

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

In Human-Computer Interaction

Citations in the corpus

Histogram of Oriented Principal Components for Cross-View Action Recognition (2014, background): “Keywords: Spatio-temporal keypoint, pointcloud, view invariance.”
HOPC: Histogram of Oriented Principal Components of 3D Pointclouds for Action Recognition (2014, background): “Another approach is to find a view independent latent space [7,8,12,14] in which features extracted from the actions captured at different view points are directly comparable.”
Cross-View Action Modeling, Learning, and Recognition (2014, background): “Another possible approach is to interpolate across views via transfer learning [13].”
View Adaptive Recurrent Neural Networks for High Performance Human Action Recognition from Skeleton Data (2017, background): “Researchers have paid much attention to this issue and designed view-invariant approaches for action recognition from color videos [16, 30, 3, 32, 20, 7, 33, 41, 23, 15, 22, 43, 44, 25, 49, 29, 8].”
Discriminative virtual views for cross-view action recognition (2012, methodology): “Once the optimal transformations are computed, we compute cross-view features x̂ for all training samples and use the subset of labeled samples to train a cross-view action classifier.”
Reshaping Visual Datasets for Domain Adaptation (2013, background): “The first column shows five groups of training datasets, each being a different view, denoted by A,B and C.”
Cross-view activity recognition using Hankelets (2012, methodology): “The results of the experiments are summarized in Table 3 where the rows and columns correspond to training (source) and testing (target) views, respectively, and the columns Ours and A show the average accuracy using BoHks and cuboids without model transfer as reported in [16], respectively.”
Intelligent multi-camera video surveillance: A review (2013, background): “Therefore, some higher level features which can be shared across camera views are further learned in (Liu et al., 2011).”
Learning View-Invariant Sparse Representations for Cross-View Action Recognition (2013, methodology): “[20] used a bipartite graph to model the relationship between two view-dependent codebooks.”
Latent Multitask Learning for View-Invariant Action Recognition (2013, background): “They either transform view-dependent video features to a new view-invariant feature space [11, 12], or adapt model parameters to the target views [3, 4, 26].”
Harnessing Lab Knowledge for Real-World Action Recognition (2014, methodology): “Transfer learning for action recognition has mainly focused on cross-view action recognition, e.g., Liu et al. (2011), Farhadi and Tabrizi (2008).”
Subspace Interpolation via Dictionary Learning for Unsupervised Domain Adaptation (2013, background): “Human actions are recognized from an unseen target view using training data taken from source views [21, 20].”
3D Object Detection with Multiple Kinects (2012, unknown): “Multiple camera views have also been used for marker-less motion capture, e.g. [13] is able to distinguish two interacting people in a studio (green box) scenario.”
View Adaptive Neural Networks for High Performance Skeleton-Based Human Action Recognition (2018, background): “Some attempts have been made in previous works to overcome the view variation problem for robust action recognition [3], [8], [9], [14], [15], [19], [24], [26], [31], [36], [37], [39], [40], [50], [52], [53], [60].”
Recognizing Micro-Actions and Reactions from Paired Egocentric Videos (2016, background): “Another relevant task in which multiple videos are used is cross-view action recognition [12, 19, 41], where they focus on the variations in appearances of actions in accordance with the changes in”
Learning a Deep Model for Human Action Recognition from Novel Viewpoints (2016, methodology): “Besides the limitations of employing linear transformation functions between views, existing cross-view action recognition methods [15], [18], [20], [21], [22], [23] are either not”