IRJET- A Survey on the Enhancement of Video Action Recognition using Semi-Supervised Learning

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 06 Issue: 04 | Apr 2019 www.irjet.net p-ISSN: 2395-0072
© 2019, IRJET | Impact Factor value: 7.211 | ISO 9001:2008 Certified Journal | Page 5149
A SURVEY ON THE ENHANCEMENT OF VIDEO ACTION RECOGNITION
USING SEMI-SUPERVISED LEARNING
Saniya M Sunil1, Dileep V K2
1Saniya M Sunil, M.Tech Student, Department of Computer Science and Engineering, LBS Institute Of technology
For Women, Kerala, India
2Dileep V K , Assistant Professor, Department of Computer Science and Engineering, LBS Institute Of technology
For Women, Kerala, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - Many efforts has been put forward to achieve
the performances of human activity or the gesture
recognition in videos by the transformation of action
knowledge fetched from the still images to videos. In this
paper, an adaptation method has been used to transform
activity or the action recognition in videos by adapting
knowledge from images. The adapted knowledge is used
to learn the mutual related semantic actions by inquiring
the common elements of both labelled videos and images.
The existing action recognition method use supervised
method for action recognition so that it is very difficult to
collect the labelled videos that cover different types of
actions. In such situation, over fitting would be an inherent
problem and the performance of activity recognition is
confined and becomes excessively complicated. Thus, the
overfitting can be mitigated and the performance of
human activity recognition is improved. Meanwhile, we
expand the adaptation method to a semi-supervised
framework which can both labelled and unlabeled videos.
KeyWords: Semi-supervised learning, Action
recognition, Knowledge adaptation, labelled and
unlabelled videos.
1. INTRODUCTION
With the fast progress of Internet and mobile phone,
activity acknowledgment in individual recordings has
turned into an essential research point because of its wide
applications such as automatic video tracking, video
annotation, video explanation, and so on. Recordings
which are uploaded on the web by the users are
transferred by clients and created by handy cameras may
contain extensive camera shake and disturbances,
hindrance, and jumbled foundation. Therefore, these
recordings contain huge intra class variations within the
same category in this manner. Hence, altogether now it is a
challenging job to recognize human actions in such similar
videos. A large number of local or confined features,
motion scale invariant feature transform are extracted
from videos then and all local features are quantized into a
histogram vector using bag-of-words illustration. Then the
vector-based classifiers are finally used to perform the
action recognition in testing videos. When the videos are
entirely simple, recognition methods have accomplished
promising results. However, noises and uncorrelated
information may be incorporated into the bow during the
extraction and quantization of the local features.
Therefore, we can come to an idea that these methods are
generally not robust. and cannot be generalized well when
the videos contain specific camera shake, occlusion,
cluttered background and so on.
In order to attain recognition accuracy, meaningful
elements of actions such as related objects, human
gestures, behavior etc should be applied to form a clearer
semantic understanding of human actions. The
effectiveness of leveraging related object or human poses
or actions have been demonstrated in recent efforts. The
methods may require to training process with huge
amounts of videos to obtained good performance,
especially for real world videos. Though, it is really
challenging to collect enough labelled videos that cover a
distinct range of action poses. Knowledge alteration or
adjustment from images to videos have exhibited
improved performance in application areas of cross media
recognition and retrieval. Knowledge adaptation is also
known as transfer learning in which the target is to
disseminate the knowledge from ancillary domains to
target domains.
2. ACTION RECOGNITION ENHANCEMENT
METHODS
A. Pose Primitive Based Human Action Recognition in Videos
or Still Images.
In 2008, Christian Thurau, et. al proposed this
paper. In this paper, they have presented a pose based
approach for action recognition from still images and
image sequences. This approach does not involve any
background subtraction or a still camera and can be
certainly extended to multiple persons. In the learning
mode, parameter representing poses and actions are
evaluated from videos. On the other hand, the method can
be used for both videos and still images in the run mode.
For recognizing pose, they used Histogram of Oriented
Gradient (HOG) based descriptor to better contend with

attached poses and cluttered or disordered background.
Actions are represented using HOG and the action
recognition is done based on a simple histogram
comparison. The proposed approach does not confide on
background subtraction or dynamic features. Therefore
allows for action recognition in still images.
B. Evaluating Color Descriptors for Object and Scene
Recognition
K. E. Van De Sande, et. al presented this paper in
2010. The paper literally describes the invariance or
consistent properties and the discreteness of color
descriptors in a ordered way. The invariable properties of
color descriptors are exhibited logically using a taxonomy
or allocation based on invariable properties with respect
to the photometric transformations. The discreteness of
color descriptors can be assessed analytically using two
benchmarks from the image domain as well as the video
domain. From the hypothetical results, any changes in the
light intensity, the advantage of invariance is category-
specific.
C. Grouplet: A structured image representation for
recognizing human and object interactions.
In 2010, B. Yao et. al propounded this paper. They
presented a novel image feature representation concept
called the “grouplet”. Grouplet defines a small group which
captures the structured or a disciplined information of an
image by encoding or encrypting a number of
differentiative visual characters and their spatial
configurations. By applying a dataset of seven different
people-playing-musical-instrument (PPMI) activities, they
demonstrated that the grouplets are more efficient for the
classification and detection of human-object interactions
compared to the other state-of-the-art methods. For
example, consider an example people-playing-musical-
instrument (PPMI); that distinguishes an individual
playing violin from an individual holding a violin needs
elegant prominence of peculiar image features and the
feature arrangements that distinguish these two scenes.
D. Multimedia Retrieval Framework Based on Semi-
Supervised Ranking And Relevance Feedback.
In 2012 Y. Yanghas has been worked on this paper
and the paper describes two types of algorithm which is a
local regression and global alignment algorithm to learn a
robust Laplacian matrix for ranking or grading. The local
linear regression method is used to anticipate the ranking
scores of its neighboring points whereas the semi
supervised are used for long term relevance feedback(RF)
algorithm to reline the multimedia input data
representation in the multimedia attribute space and the
history relevance feedback information provided by users.
And they are applied to various content-based multimedia
retrieval application which include cross media retrieval
and 3D motion or pose data retrieval. It is used on four
data sets have demonstrated its advantage in preciseness,
robustness, flexibility, and computational efficiency. The
proposed framework attains noteworthy performance.
E. Visual event recognition in videos by learning from Web
data.
This paper has been worked by L. Duan, et. al in
2012. They have presented a visual event recognition
framework which can be used for the consumer videos by
authorizing a huge quantity of loosely or freely labeled
web videos such as from YouTube website. Though the
event recognition is extremely a challenging computer
vision task, it has attracted growing attention because of
the increasingly critical demand on and recovering from a
huge amount of unconfined consumer videos. A new
pyramid matching method called ASTPM which means the
Aligned-Space-Time-Pyramid-Matching algorithm and a
novel based transfer learning method which is referred to
as A-MKL is depicted here to be better fuse the
information from multiple pyramid levels and the
different types of confined features.
F. Dense Trajectories and Motion Boundary Descriptors for
Action Recognition.
H. Wang, et. al propounded this paper in 2013.
This paper defines the video illustration which are based
on the dense trajectories and the motion boundary
descriptors. Trajectories(path) can capture or it will get
the local gesture information or facts of the videos. A
dense representation assures a proper handling of
foreground motion and also the surrounding context. The
state-of-art optical flow algorithm provides a robust and
effective extraction of the dense trajectories. illustration
and also could be applied in the circumstances of action
localization and the video retrieval.
G. Learning Discriminative Key Poses for Action Recognition.
In 2013, L. Liu, et. al has been worked on this
paper. A new method have been presented based on the
key poses representation to represent The poses in the
video frames are distinguished by the proposed extensive
pyramidal features which can include the feature maps
like Gabon, Gaussian, and wavelet pyramids. All these
features are used to encode the orientation, intensity and
configuration information and thereby providing an
informative representation of the human poses or actions.
Here to learn the subset of various discriminative poses
for representing actions, we utilize the Ada boost
algorithm.
H. Action recognition using non-negative action component
representation and sparse basis selection.

In 2014, H. Wang proposed this paper by using the higher
level action units for representing the human actions in
videos and then depending upon those units, a novel
sparse based model has been developed for the human
action recognition. In this method, they have been
introduced three interconnected components. At first, they
has been presented a new context aware spatial and a
temporal descriptor for enhancing the bias of the locally
spatial as well as the temporal descriptors which was
applied traditionally. Secondly, they attempted to study
understand or the action units by using the graph
equalized positive matrix factorization from the statistics
of the context-aware descriptors. This will leads to the
part-based illustration and also encodes or conceals the
geometrical information. These units effectively connect
the acceptable gap in human action recognition. Thirdly,
we defines a sparse model based on a joint L2,1-norm
form which has been proposed to perform the feature
selection to preserve the representative components and
conceal the noise in the action units.
I. Learning with augmented features for heterogeneous
domain adaptation.
L. Duan, et. proposed this paper in 2014. They
have been introduced a modern technique for HDA which
means Heterogeneous-Domain-Adaptation. In HFA, they
augmented the heterogeneous features from the source as
well as the target domains by using two newly proposed
feature mapping functions. They first proposed to find the
two projection matrices for the source as well as the target
data by applying the standard learning algorithm Support
Vector Machine with the hinge loss in both linear and the
nonlinear circumstances.
J. Knowledge adaptation with partially shared features for
event detection using few exemplars.
In 2014, Z. Ma presented this paper. They
introduced the research exploration of Multimedia event
detection along with few exemplars. Since this focuses on
more generic, complicated and meaningful events, this is
an important research issue that reﬂects in our daily
activities. In addition, the situation we have faced in the
real-world needs that only few positive examples are used.
For achieving great performance, we have proposed to
acquire strength from the available concepts-based videos
for the MED with little exemplars.
K. Multilevel Chinese Takeaway Process and Label based
Process For Rule Induction in the context of Automated
Sports Video Annotation.
In 2014, A. Khan, et. al worked this paper and the
paper propose four alternatives of a hierarchical hidden
Markov model strategy for indication in the frame of
reference of automated sport video annotation comprising
a Multilevel-Chinese-Takeaway; which is a process and a
recent technique known as the Cartesian-product-label-
based-hierarchical bottom-up clustering; which is called
CLHBC method that utilize previous information
incorporated within the label design. Optimal performance
is acquired using the hybrid topological arrangement with
CLHBC generated case labels.
L. Semi Supervised Features Selection via Spline Regression
for Video Semantic Recognition.
In 2015, Y. Han proposed a paper which issued to
enhance both the efficiency and accuracy of the video
semantic action recognition, it can perform feature
selection on the derived video features to select the subset
of characters form the high dimensional feature set for a
dense and exact video data representation. This discloses
semi-supervised attribute selection algorithms to better
recognize the appropriate video features, which are
biasive target classes by effectively utilizing the
information underlying the large quantity of unlabeled
video data.
M. Kernelized multiview projection for robust action
recognition.
This method has been intended by L. Shao, et. al in
2015. In this paper, they have introduced an efficient sub-
space training framework based on KMP for human action
or gesture recognition. KMP can encrypt a different kinds
of features in different ways to attain a semantically
significant embedding. A relevant feature of KMP is that it
is able to effectively explore the equivalent property of
distinct views and eventually detects a unique low-
dimensional subspace where the distribution of each view
is adequately smooth and differentiative.
N. Learning Spatio-Temporal Representations For Action
Recognition: A Genetic Programming Approach.
In 2016, L. Liu, et. al proposed an article based on
the Genetic programming approach. In this article, instead
of applying handmade features, we instinctively learn the
spatial as well as temporal motion characters or features
for action recognition. This is used to achieve via a
generative evolutionary technique, that means the Genetic
Programming, an automatic method which derives the
motion or the gesture feature descriptor structure on a
community of primitive or elementary 3-D operators. The
GP evolving feature selection or derivation methods is
estimated on four common action datasets, particularly
KTH, HMDB51,UCF. This evolves an adaptive learning style
using GP (Genetic Programming) to evolve biasive spatio-
temporal illustration, which all together fuse the color and
motion data or knowledge for high-level action
recognition tasks.
O. Structure Preserving Binary Representations for RGB-D
Action Recognition.

In 2016, M. Yu, et. al has been worked on this
paper which aims on the Local representation for RGB-D
(which is generally a combination of RGB image and its
depth information) video data fusion with a structure or
an arrangement preserving projection. For obtaining a
general characteristic for the video data, we convert the
problem to defining the gradient fields of RGB and depth
information of video sequences. With the local click
alterations or the changes of the gradient fields, which
comprise the orientation and magnitude of the
neighborhood of each point, a new kind of continuous
local descriptor called local flux feature is acquired. This
will obtain a fused local a binary illustration for RGB-D
human action recognition.
P. Discriminative Tracking Using Tensor Pooling.
In 2016, B. Ma, et. al proposed a paper to
represent target templates and candidates directly with
sparse coding tensor. Local sparse representation has
been successfully applied to visual tracking, owing to its
discriminative nature and robustness against local noise
and partial occlusions. Local sparse codes computed with
a template actually constitute a three-order tensor
according to their original layout, although most existing
pooling operators convert the codes to a vector by
concatenating or computing where it is used to deliver
more informative and structured information, which
potentially enhances the discriminative power of the
appearances model and improves the tracking
performances.
Q. Transfer Latent SVM for joint Recognition and
Localization of Actions in videos.
In 2016, C. Li, et. al has been worked on this paper
and it is based on web images and weakly annotated
training videos. The model takes training videos which are
only annotated with action label as input for alleviating
the laborious and time-consuming manual annotations of
action locations. For the purpose of improving the
localization we collect an number of web images which are
annotated with both action labels and action location to
learn a discriminative model by enforcing the local
similarities between videos and web images.
R. Multi Surface Analysis for Human Action Recognition in
Video.
This paper has been presented by Hong-Bo Zhang,
et. al in 2016. They proposed a novel multi-surface feature
named 3SMF. The prior probability is estimated by an
SVM, and the posterior probability is computed by the
NBNN algorithm with STIP. We model the relationship
score between each video and action as a probability
inference to bridge the feature descriptors and action
categories. The main contributions of our study is that a
new holistic feature (3SMF) is proposed to represent
video. 3SMF can reflect the difference of the action in
different surfaces.
S. The Comparison between the SIFT and SURF.
The paper has been proposed by Darshana Mistry
and Asian Banerjee in 2017. SIFT is used for finding
uniqueness features. In this paper it is said that the SURF
is three times better than that of SIFT because of using the
integral image and box filters. Here the SIFT will take
more time to extract the features when compared to SURF.
SIFT and SURF, both are robust method inorder to find
f e a t u r e d e t e c t i o n a n d m a t c h i n g .
T. Label Information Guided Graph Construction For Semi-
Supervised Learning.
In 2017, B. Ma, L. Huang, J. Shen and L. Shao
proposed a paper based on semi-supervised learning
methods which use the label information of observe
sample in the label propagation stage, while ignoring such
valuable information when learning the graph. The
enforcing the weight of edges between labeled samples of
different classes to the state of the art graph learning
methods, such as the low rank representation learning
method called semi supervised low rank representation.
3. CONCLUSION
To achieve the overall performance of action
recognition, we propose a classifier of Image to Video
Adaptation, which is able to acquire the knowledge from
images based on commonly visual features. At the same
time, it can completely utilize the heterogeneous features
of unlabeled videos for improving the performance of
human action recognition in the videos. Empirical results
reveal that the knowledge learned from the images can
make an impact in the recognition accuracy or fidelity of
the videos. Moreover, the results prove that the intended
IVA(Image to video adaptation) exhibit the improved
performances of human action recognition.
REFERENCES
[1] Christian Thurau, Vaclav hlavac, “Pose primitive based
human action recognition in videos or still images”,
Proceedings of the conference of computer vision and
pattern recognition, Anchorage, Alaska, USA, June
2008.
[2] K. E. Van De Sande, T.Gevers, and C. G. Snoek,
“Evaluating color descriptors for object and scene
recognition,” IEEE Trans. Pattern Analysis, vol. 32, no.
9, pp. 1582–1596, Sep. 2010.
[3] Bangpeng Yao, Li Fei-Fei, “Grouplet: A structured
image representation for recognizing human and
object interactions”, IEEE Computer Society

Conference on Computer Vision and Pattern
Recognition, 2010.
[4] Yi Yang, Feiping Nie, Dong Xu, Jiebo Luo, Yueting
Zhuang, and Yunhe Pan, “A Multimedia Retrieval
Framework Based on Semi-Supervised Ranking and
Relevance Feedback”, IEEE transactions on pattern
analysis & machine intelligence, vol 34, No.4, April
2012.
[5] Lixin Duan, Dong Xu, Ivor Wai-Hung Tsang, and Jiebo
Luo, “Visual Event Recognition in Videos by Learning
from Web Data”, IEEE transactions on pattern analysis
& machine intelligence, VOL. 34, NO. 9, SEPTEMBER
2012.
[6] Heng Wang, Alexander Klaser, Cordelia Schmid,
Cheng-Lin Liu,, “Dense Trajectories and Motion
Boundary Descriptors for Action Recognition”,
International Journal of Computer Vision, Springer
Verlag, 2013.
[7] Li Liu, Ling Shao, Xiantong Zhen, and Xuelong Li,
“Learning Discriminative Key Poses for Action
Recognition”, IEEE TRANSACTIONS ON CYBERNETICS,
VOL. 43, NO. 6, December 2013.
[8] Haoran Wang, Chunfeng Yuan, Weiming Hu, Haibin
Ling, Wankou Yang, and Changyin Sun, “Action
Recognition Using Non-negative Action Component
Representation and Sparse Basis Selection”, IEEE
TRANSACTIONS ON IMAGE PROCESSING, VOL. 23, NO.
2, February 2014.
[9] Lixin Duan, Dong Xu, and Ivor W. Tsang, “Learning
with Augmented Features for Heterogeneous Domain
Adaptation”, IEEE transactions on pattern analysis &
machine intelligence, VOL. 36, NO. 6, June 2014.
[10] Zhigang Ma, Yi Yang, Nicu Sebe, and Alexander G.
Hauptmann, “Knowledge Adaptation with Partially
Shared Features for Event Detection Using Few
Exemplars”, IEEE transactions on pattern analysis &
machine intelligence, VOL. 36, NO. 9, September 2014.
[11] A. Khan, D. Windridge, and J. Kittler, “Multi-Level
Chinese Takeaway Process and Label-Based Processes
for Rule Induction in the Context of Automated Sports
Video Annotation”, IEEE TRANSACTIONS ON
CYBERNETICS, October 2014.
[12] Yahong Han, Yi Yang, Zhigang Ma, Yan Yan, Nicu Sebe,
Xiaofang Zhou, “Semi-Supervised Feature Selection via
Spline Regression for Video Semantic Recognition”,
MANUSCRIPT SUBMITTED TO IEEE TRANSACTIONS
ON NEURAL NETWORKS AND LEARNING SYSTEMS,
February 2015.
[13] L. Shao, L. Liu, and M. Yu, “Kernelized multiview
projection for robust action
recognition”, International Journal of Computer
Vision · October 2015.
[14] Li Liu, Ling Shao, Xuelong Li, and Ke Lu, “Learning
Spatio-Temporal Representations for Action
Recognition: A Genetic Programming Approach”, IEEE
TRANSACTIONS ON CYBERNETICS, VOL. 46, NO. 1,
January 2016.
[15] Mengyang Yu, Li Liu, and Ling Shao, “Structure-
Preserving Binary Representations for RGB-D Action
Recognition”, IEEE transactions on pattern analysis &
machine intelligence, VOL. 38, NO. 8, August 2016.
[16] Bo Ma, Lianghua Huang, Jianbing Shen, and Ling Shao,
“Discriminative Tracking Using Tensor Pooling”, IEEE
TRANSACTIONS ON CYBERNETICS, VOL. 46, NO. 11,
November 2016.
[17] Cuiwei Liu, Xinxiao Wu, and Yunde Jia, “Transfer
Latent SVM for Joint Recognition and Localization of
Actions in Videos”, IEEE TRANSACTIONS ON
CYBERNETICS, VOL. 46, NO. 11, November 2016.
[18] Hong-Bo Zhang, Qing Lei, Bi-Neng Zhong, Ji-Xiang Du,
Jialin Peng, Tsung-Chih Hsiao and Duan-Sheng Chen,
“Multi-surface analysis for human action recognition
in video”, Zhang et al. SpringerPlus, 2016.
[19] Darshana Mistry and Asim Banerjee, “Comparison of
Feature Detection and Matching Approaches: SIFT and
SURF”, GRD Journals- Global Research and
Development Journal for Engineering, Volume 2, Issue
4, March 2017.
[20] Liansheng Zhuang, Zihan Zhou, Shenghua Gao,
Jingwen Yin, Zhouchen Lin and Yi Ma, “Label
Information Guided Graph Construction for Semi-
Supervised Learning”, IEEE transactions on image
processing, VOL. 26, NO. 9, September 2017.

IRJET- A Survey on the Enhancement of Video Action Recognition using Semi-Supervised Learning

More Related Content

What's hot (20)

Similar to IRJET- A Survey on the Enhancement of Video Action Recognition using Semi-Supervised Learning (20)

More from IRJET Journal (20)

Recently uploaded (20)

IRJET- A Survey on the Enhancement of Video Action Recognition using Semi-Supervised Learning