Recognize human activities from multi-part missing videos

Xu, Kaiping; Wang, Guolong

doi:10.1109/icme.2016.7552941

Cited by 4 publications

(10 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…SVM classifier is the most used handcrafted classifier because it offers a very high rate of recognition performance and classification. It was used with a Gaussian, Radial basis function (RBF) or linear kernel with or without parameters in [21,27,124,191,222,223]. (b) Naive Bayesian classifier: This is a probabilistic classifier based on the theorem of Bayes [10].…”

Section: I) First Stage Methods (Detection)mentioning

confidence: 99%

“…In [191], a comparison between this classifier and SVM is carried out. (c) Algorithm of K Nearest Neighbor: it consists in the classification of objects based on the majority of their neighbors' vote [21,222]. To identify neighbours, objects are represented by position vectors in the multidimensional space of features.…”

Section: I) First Stage Methods (Detection)mentioning

confidence: 99%

“…The work in [198] compares the obtained classification results using this algorithm and SVM. (d) K means: This is a clustering algorithm, which is sensitive to data structures and consists to iteratively calculate the k-distances from each class centroid to each datum [222]. The points are then assigned to the nearest cluster and the centers are re-evaluated to be the average of their class values.…”

Section: I) First Stage Methods (Detection)mentioning

confidence: 99%

“…HAR has been widely used for various applications such as human-computer interaction systems HCI, computer vision and augmented reality [14,19,53,65,138,222,223]. It is important to stress the interest of having interactive and natural interfaces, where the user can use his performed actions to provide instructions to the machine.…”

Section: Har Applicationsmentioning

confidence: 99%

“…However, the use of over-complete dictionaries proves to be more interesting because they can produce even more compact representations. Therefore, the authors of [222] have combined these two concepts to propose a solution of HAR from multi-part missing video. (b) Genetic programming (GP): GP is an evolutionary method inspired by the process of biological evolution and its fundamental mechanism [24].…”

Section: Har Approaches According To Feature Extraction Processmentioning

confidence: 99%

See 4 more Smart Citations

Vision-based human activity recognition: a survey

Beddiar

Nini

Sabokrou

et al. 2020

Multimed Tools Appl

330

117

View full text Add to dashboard Cite

Human activity recognition (HAR) systems attempt to automatically identify and analyze human activities using acquired information from various types of sensors. Although several extensive review papers have already been published in the general HAR topics, the growing technologies in the field as well as the multi-disciplinary nature of HAR prompt the need for constant updates in the field. In this respect, this paper attempts to review and summarize the progress of HAR systems from the computer vision perspective. Indeed, most computer vision applications such as human computer interaction, virtual reality, security, video surveillance and home monitoring are highly correlated to HAR tasks. This establishes new trend and milestone in the development cycle of HAR systems. Therefore, the current survey aims to provide the reader with an up to date analysis of vision-based HAR related literature and recent progress in the field. At the same time, it will highlight the main challenges and future directions.

show abstract