Learning to Recognize Activities from the Wrong View Point

By Ali Farhadi, Mostafa Kamali Tabrizi. 2008

221 citations recorded by the source corpus.

Subfields: 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 Biomedical Engineering

In Computer Vision and Pattern Recognition

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, methodology): “Transfer learning-bsaed cross-view action recognition methods [3, 11, 25] are based on local appearance features.”
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].”
Efficient Learning of Domain-invariant Image Representations (2013, methodology): “Recently proposed feature adaptation methods [1, 2, 12, 13, 14, 15] offer a solution by learning a category-independent feature transform that maps target features into the source, pooling all training labels across categories.”
Discriminative virtual views for cross-view action recognition (2012, methodology): “See [5, 25] for more details.”
A latent model of discriminative aspect (2009, background): “Features consist of three channels: smoothed horizontal flow, smoothed vertical flow, and the silhouette.”
Cross-View Action Recognition via Transferable Dictionary Learning (2016, methodology): “generated the same split-based features for correspondence video frames from both training and testing views [11].”
A Comprehensive Study on Cross-View Gait Based Human Identification with Deep CNNs (2017, background): “For example, action recognition rates on the INRIA Xmas Motion Acquisition Sequences dataset [3] can drop from 80.0 percent [3] to lower than 20.0 percent [4], when evaluated under cross-view conditions.”
Discovering Latent Domains for Multisource Domain Adaptation (2012, background): “Other work on feature-translation has included translating features between camera views to transfer activity models [13], and translating user preferences between text and image domains [14].”
Cross-View Action Recognition over Heterogeneous Feature Spaces (2013, methodology): “From the perspective of cross-view action recognition, some work [2, 9, 20] is closely related to our approach.”
Transfer Learning for Activity Recognition: A Survey (2013, methodology): “[20] Video camera View angle US Feature-space”
Learning View-Invariant Sparse Representations for Cross-View Action Recognition (2013, methodology): “Instead of transferring the split-features as in [6], we transfer the indices of the non-zero elements (i.”
Harnessing Lab Knowledge for Real-World Action Recognition (2014, methodology): “Similarly, the method in Farhadi and Tabrizi (2008) also deals with cross-view action recognition and is restrained to one specific multi-view dataset.”
Cross-view activity recognition using Hankelets (2012, methodology): “Table 5 gives side by side the average accuracy using BoBHks with the accuracy of previous approaches [16, 7, 10, 6].”
View-Invariant Action Recognition Using Latent Kernelized Structural SVM (2012, methodology): “Farhadi and Tabrizi [13] used Maximum Margin Clustering to generate split features in the source view and then learned the split features in the target view by the transferred split values from source view.”
Cross-view action recognition via view knowledge transfer (2011, methodology): “We are particularly interested in [2], since they also transfer action models across views.”
Reshaping Visual Datasets for Domain Adaptation (2013, background): “(In contrast, test sets for the object recognition datasets above are less heterogeneous.)”