Action recognition in videos acquired by a moving camera using motion decomposition of Lagrangian particle trajectories

Wu, Shandong; Oreifej, Omar; Shah, Mubarak

doi:10.1109/iccv.2011.6126397

Cited by 132 publications

(91 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Gilbert et al [34] describes a hierarchical representation of features in a multi-stage approach to capture the most distinctive components of actions. Wu et al [35] uses optical flow to find motion, thereby avoiding the need for object detection and creating robust trajectories. These motion features are then used in an SVM for performing action recognition.…”

Section: Related Workmentioning

confidence: 99%

Action recognition using global spatio-temporal features derived from sparse representations

Somasundaram

Cherian

Morellas

et al. 2014

Computer Vision and Image Understanding

View full text Add to dashboard Cite

Recognizing actions is one of the important challenges in computer vision with respect to video data, with applications to surveillance, diagnostics of mental disorders, and video retrieval. Compared to other data modalities such as documents and images, processing video data demands orders of magnitude higher computational and storage resources. One way to alleviate this difficulty is to focus the computations to informative (salient) regions of the video. In this paper, we propose a novel global spatio-temporal selfsimilarity measure to score saliency using the ideas of dictionary learning and sparse coding. In contrast to existing methods that use local spatio-temporal feature detectors along with descriptors (such as HOG, HOG3D, HOF, etc.), dictionary learning helps consider the saliency in a global setting (on the entire video) in a computationally efficient way. We consider only a small percentage of the most salient (least self-similar) regions found using our algorithm, over which spatio-temporal descriptors such as HOG and region covariance descriptors are computed. The ensemble of such block descriptors in a bag-of-features framework provides a holistic description of the motion sequence which can be used in a classification setting. Experiments on several benchmark datasets in video based action classification demonstrate that our approach performs competitively to the state of the art.

show abstract

Section: Related Workmentioning

confidence: 99%

Action recognition using global spatio-temporal features derived from sparse representations

Somasundaram

Cherian

Morellas

et al. 2014

Computer Vision and Image Understanding

View full text Add to dashboard Cite

show abstract

“…Since our focus here is on modeling a spatiotemporal graph of video features -not on particular features -we use two recent, easy-to-implement approaches that were thoroughly evaluated in the literature [1,18].…”

Section: Feature Extractionmentioning

confidence: 99%

“…Feature extraction: Given a video, motion features are extracted in the form of trajectory snippets, which can be either KLT tracks of Harris corners [1], or Lagrangian particle trajectories [18]. These features have been demonstrated by prior work as robust against many challenges of real-world videos, including large variations in camera motions, object pose and scale, and illumination.…”

Section: Introductionmentioning

confidence: 99%

Human Activities as Stochastic Kronecker Graphs

Todorovic

2012

Computer Vision – ECCV 2012

View full text Add to dashboard Cite

Abstract.A human activity can be viewed as a space-time repetition of activity primitives. Both instances of the primitives, and their repetition are stochastic. They can be modeled by a generative model-graph, where nodes correspond to the primitives, and the graph's adjacency matrix encodes their affinities for probabilistic grouping into observable video features. When a video of the activity is represented by a graph capturing the space-time layout of video features, such a video graph can be viewed as probabilistically sampled from the activity's modelgraph. This sampling is formulated as a successive Kronecker multiplication of the model's affinity matrix. The resulting Kronecker-power matrix is taken as a noisy permutation of the adjacency matrix of the video graph. The paper presents our: 1) model-graph; 2) memory-and time-efficient, weakly supervised learning of activity primitives and their affinities; and 3) inference aimed at finding the best expected correspondences between the primitives and observed video features. Our results demonstrate good scalability on UCF50, and superior performance to that of the state of the art on individual, structured, and collective activities of UCF YouTube, Olympic, and Collective datasets.

show abstract

“…Sun et al [4] used trajectories of SIFT points and encoded motion in three levels of context information: point level, intra-trajectory context and inter-trajectory context. Wu et al [5] used a dense trajectory field obtained by tracking densely sampled particles driven by optical flow. They decomposed the trajectories into camera-induced and object-induced components using low rank optimisation.…”

Section: Introductionmentioning

confidence: 99%

Motion Trend Patterns for Action Modelling and Recognition

Nguyễn

Manzanera

Garrigues

2013

Computer Analysis of Images and Patterns

View full text Add to dashboard Cite

Abstract.A new method for action modelling is proposed, which combines the trajectory beam obtained by semi-dense point tracking and a local binary trend description inspired from the Local Binary Patterns (LBP). The semi dense trajectory approach represents a good trade-off between reliability and density of the motion field, whereas the LBP component allows to capture relevant elementary motion elements along each trajectory, which are encoded into mixed descriptors called Motion Trend Patterns (MTP). The combination of those two fast operators allows a real-time, on line computation of the action descriptors, composed of space-time blockwise histograms of MTP values, which are classified using a fast SVM classifier. An encoding scheme is proposed and compared with the state-of-the-art through an evaluation performed on two academic action video datasets.

show abstract

Action recognition in videos acquired by a moving camera using motion decomposition of Lagrangian particle trajectories

Cited by 132 publications

References 25 publications

Action recognition using global spatio-temporal features derived from sparse representations

Action recognition using global spatio-temporal features derived from sparse representations

Human Activities as Stochastic Kronecker Graphs

Motion Trend Patterns for Action Modelling and Recognition

Contact Info

Product

Resources

About