SIFT-Bag kernel for video event analysis

By Xi Zhou, Xiaodan Zhuang, Shuicheng Yan, Shih-Fu Chang, Mark Hasegawa-Johnson, Thomas S. Huang. 2008

140 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

3D Human Activity Recognition with Reconfigurable Convolutional Neural Networks (2014, background): “A batch of works on human action/activity understanding mainly focused on developing robust and descriptive features [35, 6, 20, 19, 40, 37, 24].”
Metric Learning for Large Scale Image Classification: Generalizing to New Classes at Near-Zero Cost (2012, unknown): “The second is the nearest class mean classifier (NCM), where classes are represented by their mean feature vector [13,14,15].”
Cross-dataset action detection (2010, methodology): “Following this approach, a lot of recent work employs GMM to model the temporal structure of audio-visual events [23] or the visual similarities between two images or videos [24].”
A Deep Structured Model with Radius–Margin Bound for 3D Human Activity Recognition (2015, background): “Many works on human action/activity recognition mainly focus on designing robust and descriptive features [49, 13, 30, 29, 55, 51, 34].”
Learning latent spatio-temporal compositional model for human action recognition (2013, background): “Although many research studies [45, 41, 19, 46, 23, 5, 30, 37, 32, 8] have been carried out to understand and retrieve large scale video contents, these works focus on high level semantics instead of describing human activities.”
Distance-Based Image Classification: Generalizing to New Classes at Near-Zero Cost (2013, unknown): “The Sift-bag kernel of [30] is also related to our method since it uses an NCM classiﬁer and an (cid:96) 2 distance in a subspace that is orthogonal to the subspace with maximum within-class variance.”
Recognizing actions using depth motion maps-based histograms of oriented gradients (2012, background): “Automatic human action recognition has many real-world applications including content-based video search, humancomputer interaction, video surveillance, health care, and etc [7-9, 12-14].”