Action Recognition with Actons

By Jun Zhu, Baoyuan Wang, Xiaokang Yang, Wenjun Zhang, Zhuowen Tu. 2013

102 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Action recognition with trajectory-pooled deep-convolutional descriptors (2015, background): “To overcome this issue, several mid-level and high-level video representations have been proposed such as Action Bank [22], Dynamic-Poselets [37], Motionlets [35], Motion Atoms and Phrases [34], and Actons [42].”
Actions ~ Transformations (2015, background): “Mid-to high-level representation are also used to model complex actions [41, 55, 36, 60, 14, 24].”
Temporal Segment Networks for Action Recognition in Videos (2017, background): “mid-level representations have been proposed for action recognition [3], [23], [47], [48], [49], [50], [51].”
Effective Active Skeleton Representation for Low Latency Human Action Recognition (2016, methodology): “In [19], it extends the maximum-bag-margin MIL (MI-SVM [20]) to learn these actons from training set.”
A Robust and Efficient Video Representation for Action Recognition (2015, unknown): “The best previous result is from (Zhu et al, 2013).”
AdaScan: Adaptive Scan Pooling in Deep Convolutional Neural Networks for Human Action Recognition in Videos (2016, methodology): “Other approaches have also used MIL and its variants to model discriminative frames in a video, with or without a temporal structure [26, 30, 8, 51, 42, 20, 27].”
Real-Time RGB-D Activity Prediction by Soft Regression (2016, background): “) and feature learning techniques (sparse coding [21], max-margin learning [22, 23], Fisher vector [19] etc.”
Multiple instance learning: A survey of problem characteristics and applications (2016, methodology): “A method based on multiple instance clustering was also proposed to discover a set of actons (sub-actions) from videos to create a mid-level representation of actions [118].”
Moving Poselets: A Discriminative and Interpretable Skeletal Motion Representation for Action Recognition (2015, background): “In [31], different from actemes, actons are built on top of the BoF representation of each volume of interest, forming a mid-level dictionary of intermediate concepts to characterize the semantic properties.”
Videos as Space-Time Region Graphs (2018, background): “For example, a large effort has been spent on manually designing the video features [22,23,24,25,26,27,28,29,30,31].”
Action Recognition with Stacked Fisher Vectors (2014, methodology): “, motionlets [35], mid-level parts [27], and actons [37]) that utilize the responses of discriminative action parts combined with low-level features perform inferior to our method (FV+SFV) with a certain margin on”
DeePM: A Deep Part-Based Model for Object Detection and Semantic Part Localization (2015, background): “…a wide range of computer vision tasks including object detection (Felzenszwalb et al., 2010; Chen et al., 2014), pose estimation (Yang & Ramanan, 2011; Chen & Yuille, 2014), semantic segmentation (Long et al., 2015; Chen et al., 2015) and action recognition (Yang et al., 2010; Zhu et al., 2013).”