Smart Sensor-Based Gesture Recognition: A Literature Review

Desklib provides past papers and solved assignments for students. This literature review explores AI-based gesture recognition in speech therapy.

Review Based Project Literature Review (Secondary Research)
Student's Name and CSU ID
Project Type Review Based Project
Project Name Review of Gesture Recognition techniques using Smart Sensors for the training of speech pathologists.
Technology Smart Sensors
Techniques Gesture Recognition
Domain Human-computer interaction for the education of students with speech impairment
1

Version 1.0 _ Week 1 (5 Journal Papers from CSU Library)
1
Reference in APA format that will be in
'Reference List'
Raheja, J. L., Chandra, M., & Chaudhary, A. (2018). 3D gesture-based real-time object selection and
recognition. Pattern Recognition Letters, 115, 14-19. Doi: https://doi.org/10.1016/j.patrec.2017.09.034
The citation that will be in the content Raheja et al. (2018)
URL of the Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www.sciencedirect.com/science/
article/abs/pii/S0167865517303549
Level of journal-Q1  Kinect sensor
 Object extraction
 Pointed object
 Pointing gesture recognition
 Skeleton tracking
The Name of the Current Solution
(Technique/ Method/ Scheme/
Algorithm/ Model/ Tool/ Framework/ ...
etc )
The Goal (Objective) of this Solution &
What is the Problem that needs to be solved
What are the components of it?
Technique/Algorithm name:
Harris-corner detection technique
Problem:
The issue in this work is due to the lack of  Object extraction and localization
2

Tools:
 Microsoft Kinect sensor
 Hue-Saturation Histogram
matching
Applied Area:
This work is applied for pointing gesture
for HCI systems.
localizing the 3D position of a user body.
Goal:
The main goal of this work is to attain real-
time constraint for estimating the real joint
location.
 Skeleton detection and tracking
 Calibration
The Process (Mechanism) of this Work; The process steps of the Technique/system
Process Steps Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1 Object extraction and localization
In this phase, the captured image is used to be
segmented and assumed that the object is within
the range of 4m from the 3D camera
The object can be localized Timing error
2 Skeleton detection and tracking
The user segmentation is supposed to be
identified in the scene where each and every
Auto tracking is started without requiring a
calibration pose
N/A
3

user’s data is unique
3 Calibration
In the last phase, the skeletal tracking will
provide the 3D location
Captures range of camera Kinect loses track to the user
Validation Criteria (Measurement Criteria)
Dependent Variable Independent Variable
System value Decided zone
systems User
Joint location Image position
Input and Output Critical Thinking: Feature of this work, and
Why (Justify)
Critical Thinking: Limitations of the research
current solution, and Why (Justify)
Input (Data) Output (View)
 The depth and
RGB map
will be
obtained from
Object zooming and
object recognition
will be made which
will be helpful in
In the words of Raheja et al. (2018), this work is
advantageous because:
 It helps in pointing the arm gestures easily
 The pointed object can be recognized for
further actuation
The limitation in this work is made when the
object is extracted as in this phase the objects are
supposed to be limited within a zone. Hence, the
whole outcome of the system is getting
affected(Raheja et al. 2018)
4

Kinect segmenting the object
on the basis of depth
information
 The objects can be selected by using hand
and system as well
(Describe the research/current solution) Evaluation Criteria How this research/current solution is valuable
for your project
What is the Future work that set by the author
in Conclusion and Future work section
As per the words of Raheja et al. (2018) in the
current solution, the detection and selection of
pointing objects are not made properly. Hence,
they approximate the pointing location of the
customer by restricting the image 3D position
which helps in managing the accuracy of 94%
over 220 gestures.
This work is effective and efficient as well because
it helps in detecting the pointing objects which help
in the training of Speech pathologists (Raheja et al.
2018)
According to Raheja et al. (2018), this work is
valuable because it helps in providing real-time
constraint data for analyzing the outcome of the
system.
In future work, the system will be made on the
basis of the color histogram approach for object
selection
Diagram/Flowchart
5

Figure 1 System methodology
Reference in APA format that will be in 'Reference List'
 Raheja, J. L., Chaudhary, A., & Maheshwari, S. (2014). Hand gesture pointing location detection. Optik, 125(3), 993-996. Doi:
6

https://doi.org/10.1016/j.ijleo.2013.07.167
 Chaudhary, A. (2018). Robust Hand Gesture Recognition for Robotic Hand Control. Springer. https://link.springer.com/book/10.1007%2F978-981-
10-4798-5
The citation that will be in the content
 (Raheja, Chaudhary & Maheshwari. 2014)
 (Chaudhary. 2018)
7

2
Reference in APA format that will be in
'Reference List'
Schipor, O. A., Vatavu, R. D., &Vanderdonckt, J. (2019). Euphoria: A Scalable, Event-driven Architecture for
Designing Interactions across Heterogeneous Devices in Smart Environments. Information and Software
Technology, 115, 14-19 Doi: https://doi.org/10.1016/j.infsof.2019.01.006
The citation that will be in the content (Schipor, Vatavu & Vanderdonckt. 2019)
URL of the Reference Level of Journal (Q1, Q2, …Qn) Keywords in this Reference
https://www-sciencedirect-
com.ezproxy.csu.edu.au/science/article/
pii/S0950584919300096
Level of journal-Q1  Context-aware computing
 Mobile computing
 Multi-device interaction
 Smart Environments
 Smart Spaces
 Software architecture
 Wearable computing
The Name of the Current Solution
(Technique/ Method/ Scheme/
Algorithm/ Model/ Tool/ Framework/ ...
etc )
The Goal (Objective) of this Solution &
What is the Problem that needs to be solved
What are the components of it?
Technique/Algorithm name:
Pattern recognition technique, Event-
Problem: The problem in this work is of the
lack of flexible interaction in the input and
 Data collection
 Emission
8

Response System technique
Tools:
 Proximity toolkit
 Interaction tool kit
Applied Area: Designing interactions
among heterogeneous devices in smart
environments
output device.
Goal:
To introduce Euphoria for enhancing the
interaction among system and user.
 Receiving
The Process (Mechanism) of this Work; The process steps of the Technique/system
Process Steps Advantage (Purpose of this step) Disadvantage (Limitation/Challenge)
1 Data collection
In this phase, the specific parameters of the
system will be collected.
High-efficiency N/A
2 Emission
The emitters will be built around the device for
orientation.
proxemic awareness Lack of energy performance
3 Receiving Detect and integrate changes N/A
9

The real-time performance will be received in
response to the system at the same time.
Validation Criteria (Measurement Criteria)
Dependent Variable Independent Variable
Interaction Training
Device corresponding pixel resolution,
Request-response Time
Input and Output Critical Thinking: Feature of this work, and
Why (Justify)
Critical Thinking: Limitations of the research
current solution, and Why (Justify)
Input (Data) Output (View)
 Message
processing
service
 Publication
center
The average response
time of 150 ms will be
made for message
exchange.
According to Schipor, Vatavu & Vanderdonckt.
(2019) following are the advantage of the work:
 It helps in reducing the computing power of
the system
 The flexibility in the interaction helps in
improving the education level of students
Schipor, Vatavu & Vanderdonckt. (2019) analyze
the limitation of the work in which the user is not
able to communicate with the system properly and
hence it affects the whole outcome.
10

(Describe the research/current solution) Evaluation Criteria How this research/current solution is valuable
for your project
What is the Future work that set by the author
in Conclusion and Future work section
Schipor, Vatavu & Vanderdonckt. (2019)
recognize the lack of interaction in the system.
Hence, they developed the Euphoria which is new
software for enhancing the interaction in the
system.
The developed system is effective as it assists in
managing the multi-device interaction in a
systematic and mannered way (Schipor, Vatavu &
Vanderdonckt. 2019).
This system helps in improving the flexibility in
interaction. In future work, the average request
response time will be done for reducing the
complexity of the system (Schipor, Vatavu &
Vanderdonckt. 2019)
Diagram/Flowchart
11

Figure 2: The multi-layer architecture of Euphoria, consisting of event producers, emitters, a processing engine, and event receivers and consumers
Reference in APA format that will be in 'Reference List'
12