6. Literature Review - (Summary)
Literature review of minimum 5 papers.
Either separate slide for each reference or Prepare a
single Slide with only one table as shown in slide
number 15
15. Literature Review
Sr
No.
Title Year Methodology Strengths Weakness
1 A Robust and Real-Time Face
Anti-spoofing Method Based on
Texture Feature Analysis.
2019 The method extracts the LBP features
from YCrCb and Gray color space.
Further, COALBPs (Co-occurrence of
Adjacent Local Binary Patterns) is
computed from Gray scale image.
These features are combined and
passed to SVM for binary
classification of input face images.
Highest accuracy
achieved among
the papers
considering hand
crafted features
techniques
Method is assessed
only on NUAA
dataset.
2 Face anti-spoofing based on color
texture analysis.
2015 The technique proposed aims to solve
the problem of face spoofing by
extracting color texture features. The
author tries to find out which color
space amongst YCrCb, RGB and HSV
can well distinguish a face into true or
fake classes by using color LBP
features extracted from each
individual channel.
Inter-database
evaluation is
done.
In case of replay-
attack dataset,
proposed method
lags behind as
compared to other
state-of-the-art
methods.
Texture Based Methods
16. Literature Review
Sr
No.
Title Year Methodology Strengths Weakness
3 Face Spoofing Detection Based on
Combining Different Color Space
Models.
2019 Input RGB face image is transformed
into YCrCb and LUV to extract LBP
features and into HSV to extract CM
(Color Moment) features. The
extracted features are cascaded and
passed to SVM for classification.
Proposed method
is simple and
efficient in terms
of computation.
Feature length is
large.
4 Face presentation attack
detection using guided scale texture.
2018 To eliminate irrelevant components,
the input image is transformed into
guided scale space. Then Guided
Scale Based Local Binary Pattern
(GS-LBP) and Local Guided
Binary Pattern (LGBP) descriptors
are used to extract texture features
which are then concatenated and
classified using SVM.
Unnecessary
noise is
minimized by
using GS-LBP.
sampling and
quantization
strategies need to be
improved
Texture Based Methods
19. References
For example (Follow given format):
1. Gnoying Feng et.al “Experimental research on vertical
axis wind turbine” IEEE school of energy and power
engineering, vol. 978,no.1, 2009.
2. G.M.Hasan Shaharirar “Design and construction of
vertical axis wind turbine” IEEE International form on
strategic technology, vol. 978, no. 1, pp 326-329, Oct
2014.
3. S.Sathiyamoorthy et.al, “Hybrid energy harvesting
using Piezoelectric materials, automatic rotational solar