Contenu connexe Similaire à Patent Landscape Report on Hand Gesture Recognition by PatSeer Pro (20) Patent Landscape Report on Hand Gesture Recognition by PatSeer Pro2. Page 2 of 24
© 2017 Gridlogics. All Rights Reserved.
HAND GESTURE RECOGNITION - OVERVIEW
Gesture recognition is the ability of a device to identify and respond to the different gestures of an
individual. Most gesture recognition technology can be 2D-based or 3D-based, working with the
help of a camera-enabled device, which is placed in front of the individual. The camera-enabled
device beams an invisible infrared light on the individual, which is reflected back to the camera and
onto a gesture recognition Integrated Chip (IC).
Gestures can originate from any bodily motion or state but commonly originate from the face or
hand. Current focuses in the field include emotion recognition from face and hand gesture
recognition. Users can use simple gestures to control or interact with devices without physically
touching them. Many approaches have been made using cameras and computer vision algorithms to
interpret sign language. However, the identification and recognition of posture, gait, proxemics, and
human behaviors is also the subject of gesture recognition techniques.
Gesture recognition can be seen as a way for computers to begin to understand human body
language, thus building a richer bridge between machines and humans than primitive text user
interfaces or even GUIs (graphical user interfaces), which still limit the majority of input to keyboard
and mouse.
Touchless user interface is an emerging type of technology in relation to gesture control. Touchless
user interface (TUI) is the process of commanding the computer via body motion and gestures
without touching a keyboard, mouse, or screen.
The ability to track a person's movements and determine what gestures they may be performing can
be achieved through various tools. The kinetic user interfaces (KUIs) are an emerging type of user
interfaces that allow users to interact with computing devices through the motion of objects and
bodies. Examples of KUIs include tangible user interfaces and motion-aware games such as Wii and
Microsoft's Kinect, and other interactive projects
In order to interpret movements of the body, one has to classify them according to common
properties and the message the movements may express. For example, in sign language each
gesture represents a word or phrase. The taxonomy that seems very appropriate for Human-
Computer Interaction has been proposed by Quek in “Toward a Vision-Based Hand Gesture
Interface". He presents several interactive gesture systems in order to capture the whole space of
the gestures:
Manipulative
Semaphoric
Conversational
Some literature differentiates 2 different approaches in gesture recognition: a 3D model based and
an appearance-based. The foremost method makes use of 3D information of key elements of the
body parts in order to obtain several important parameters, like palm position or joint angles. On
the other hand, Appearance-based systems use images or videos for direct interpretation.
3D model-based algorithms
Skeletal-based algorithms
Appearance-based models
3. Page 3 of 24
© 2017 Gridlogics. All Rights Reserved.
PATENT SEARCH STRATEGY
Using PatSeer, we searched individual publications and then collapsed them to one member per
Family (Simple Families). Patent families are a group of one or more patent applications in multiple
countries which represent the same invention.
The publications included in the report are updated as of 5th
September 2017.
Summary of Search Results:
We started with a set of 5386 records
We then used deduplication as Simple Families to get 2953 Families
We browsed through some records to remove all the records which were irrelevant to
the search topic
We flagged/rated each individual record using Search Recall™ option present in PatSeer
We then filtered, analyzed and came across some unwanted terms/Keywords
These records were then removed from the project by using the search within records
option
NOTE: All analysis in this report has been done on Simple Family (one member per family) and
so the data in the charts should be construed accordingly.
4. Page 4 of 24
© 2017 Gridlogics. All Rights Reserved.
TECHNICAL SEGMENTATION (PATENT
CATEGORIZATION)
To know more about the search strategy of this report and methodology used for patent
categorization Contact us here:
5. Page 5 of 24
© 2017 Gridlogics. All Rights Reserved.
FILING TREND
The chart below shows number of filings for hand Gesture Recognition during the last 15 years.
The chart was generated on overall records (5386 Records).
Trend analysis based on filing of priority application indicates a gradual increase in number of
applications being filed. Maximum number of patent applications (818) taking/having priority
were filed during 2012.
How we did it?
Application trend chart was generated within Quick Stats on the overall record set. The generated chart
was then converted to Earliest Priority Year.
6. Page 6 of 24
© 2017 Gridlogics. All Rights Reserved.
TOP COMPANIES
The chart below represents top companies active in Hand Gesture Recognition technology, with
a single representation from each family.
It is evident that Microsoft leads the patent count for Hand Gesture Recognition followed by
Samsung and Intel.
The chart below shows assignee innovation time line for the top 15 Assignees active in
hand gesture control
With the dots representing patent publications / filings and the blue lines indicating the
timelines between the earliest and latest filings for different assignees, one can look
into the relevance of filings for different assignees with respect to time
Intel appear to be one of the earliest assignees with filings for gesture recognition and
one that innovators pursued the longest right till present date. Though there are long
gaps in the timeline from the early days, the recent years have continued to see a lot of
IP activity. Similarly, Audi has visibly shorter timeline with no new patents since 2015
7. Page 7 of 24
© 2017 Gridlogics. All Rights Reserved.
How we did it?
Once the patents were added to PatSeer Projects, top assignees and innovation time line was generated
within the Quick Stats Tab.
Note:
Assignees used for all the charts below were generated with the help of custom fields, i.e. similar
assignees with name variations were filtered, and then these were assigned to a particular custom field
(assignee).This was repeated for the top assignees
8. Page 8 of 24
© 2017 Gridlogics. All Rights Reserved.
RESEARCH ACTIVITY AROUND THE WORLD
The below map represents the geographical filing relating to Gesture Recognition. The map
helps provide an indication of where innovation in this area is originating.
China is the leading country in this field with 1131 families followed by United States (919) and
Korea (321). The strength of the coloring represents the proportion of patent applications.
Country Total
CN 1131
US 919
KR 321
JP 214
TW 208
How we did it?
The map was generated using the Priority country tab under the Quick Stats option.
9. Page 9 of 24
© 2017 Gridlogics. All Rights Reserved.
COMPANIES ACTIVITY ACROSS APPLICATIONS
The table below shows research activity of companies across different applications of
hand gesture recognition
Samsung has the maximum number of records for gaming with approximately 50% of its
records concentrating on gaming
Leap Motion which is primarily into gesture recognition focuses mainly on automobile
and gaming applications
How we did it?
First records were classified under different application areas. These were then compared using the co-
occurrence tool with the companies. The resulted matrix/table chart was then exported to a CSV format
10. Page 10 of 24
© 2017 Gridlogics. All Rights Reserved.
COMPANIES ACROSS DIFFERENT TYPES SENSORS
The table below shows research activity of companies across different sensors used for
hand gesture recognition
Samsung leads the records count for Proximity, infrared and capacitive sensors
Microsoft has research activity across all the type of sensors except speed sensor
How we did it?
First records were classified under different types of sensors used for hand gesture recognition, and then
a co-occurrence matrix for assignees and sensors was generated using the co-occurrence tool. The
resulted matrix was then converted to Row Heat Map and exported.
11. Page 11 of 24
© 2017 Gridlogics. All Rights Reserved.
COMPANIES ACROSS DIFFERENT TYPES OF
GESTURES
The table below shows research activity of companies across different types of gestures
classified relating hand gesture recognition
Most of the top assignees namely Microsoft , Intel and Samsung are using dynamic
gestures
How we did it?
First records were classified under different types of gestures and then using the co-occurrence tool a
matrix was generated for the top assignees and types of gestures. The resulted matrix was then converted
to a Tree Map and exported as an image.
12. Page 12 of 24
© 2017 Gridlogics. All Rights Reserved.
COMPANIES ACROSS DIFFERENT INPUT DEVICES
The table below shows research activity of companies across different types input
devices relating hand gesture recognition
Intel and Microsoft are more into Stereoscopic and depth cameras
Hyundai motors has research activity for only stereoscopic and infrared camera which is
relative to its applications in the automobile sector
How we did it?
First records were classified under different types of input devices used for hand gesture control; matrix
was then generated for the top assignees and different input devices using the co-occurrence tool. The
resulted matrix was then converted to a Column Chart and exported as an image.
13. Page 13 of 24
© 2017 Gridlogics. All Rights Reserved.
COMPANIES ACROSS GESTURE CLASSIFICATION
ALGORITHMS
The table below shows research activity of companies across different classification
algorithms used for hand gesture recognition
Continuous Hidden Markov algorithm is the most commonly used algorithm
Qualcomm uses all the algorithms except conditional random field for its applications
relation to gesture control
How we did it?
First records were classified under different classification algorithms used for hand gesture recognition,
and then a co-occurrence matrix for assignees and those algorithms was generated using the co-
occurrence tool. The resulted matrix was then converted to Column Heat Map and exported.
14. Page 14 of 24
© 2017 Gridlogics. All Rights Reserved.
APPLICATIONS ACROSS DIFFERENT TYPES OF
SENSORS
The Heat Map below shows the different types of sensors used across various
application of hand gesture recognition
Gaming require detection of vibration, acceleration, inclination and orientation which is
possible with the help of accelerometer sensors; which are widely used for gaming
applications
Accelerometer and infrared sensors are widely used for home automation
How we did it?
First records were classified under different application areas and various sensors and then a co-
occurrence matrix for the same was generated using the co-occurrence tool. The resulted matrix was then
converted to Heat Map and exported.
15. Page 15 of 24
© 2017 Gridlogics. All Rights Reserved.
APPLICATIONS ACROSS INPUT DEVICES
The Heat Map below shows the different types of input devices used across various
application of hand gesture recognition
Stereoscopic cameras and infrared cameras are widely used across all the application
areas
Depth Cameras are used in applications like gaming, automobile and television control
to detect the relative position of users from the device.
How we did it?
First records were classified under different application areas and input devices and then a co-occurrence
matrix for the same was generated using the co-occurrence tool. The resulted matrix was then converted
to Column Chart and exported as image.
16. Page 16 of 24
© 2017 Gridlogics. All Rights Reserved.
APPLICATIONS ACROSS GESTURE CLASSIFICATION
ALOGRITHM
The chart below shows the different Cryptographic methods used for various
applications
Hidden Markov and Support Vector Machine(SVM) algorithms are widely used for all the
applications
Finite State Machine (FSM) have been used for many years in video games to model the
AI of Non Playing Characters; the chart below confirms this fact
How we did it?
First records were classified under different application areas and classification algorithms and then a co-
occurrence matrix for the same was generated using the co-occurrence tool. The resulted matrix was then
converted to a column chart and exported as an image.
17. Page 17 of 24
© 2017 Gridlogics. All Rights Reserved.
TECHNOLOGY LANDSCAPE FOR HAND GESTURE
RECOGNITION
The landscape map below represents key concepts for different companies generated across
title, abstract and claims. Themes are collection of prominent topics extracted from the patent
data and grouped under relevant parent tags.
Clusters for Touch screen and virtual reality touch screen are close to each other as there is high
degree of relevance between the records present in those themes.
How we did it?
The VizMAP tool in PatSeer Pro was used for this analysis. First the clusters were generated on title,
abstract and claims using the Advanced Clustering option. Once the clusters were generated these were
loaded using the Landscape mode and then the themes were colored by assignee.
18. Page 18 of 24
© 2017 Gridlogics. All Rights Reserved.
CITATIONS ANALYSIS FOR LEAP MOTION
Leap Motion, Inc. is an American company that manufactures and markets a computer
hardware sensor device that supports hand and finger motions as input, analogous to a mouse,
but requires no hand contact or touching. In 2016, the company released new software
designed for hand tracking in virtual reality.
Leap Motion is a hardware sensor device that tracks hand and finger motions and translates it
into input. The Leap Motion controller is a small USB powered device compatible with Windows
and Macintosh that uses two monochromatic IR cameras and three IR LEDs to track movements
and motion made by hands and fingers in a roughly 1m hemispherical 3D space. The cameras
reportedly generate 300 frames per second of data, which is analyzed and interpreted by
proprietary software
The filings by Leap Motion relate to 3Dimensional space, infrared cameras mapping hand
gestures to wearable devices and computer readable storage mediums. Many multinationals
(patenting authorities) are referring/citing these patents to evolve the technology further. The
forward citation chart below reiterates the fact. Companies like Google, Microsoft, Acer,
Samsung, Sony etc. all are citing the Leap Motion patents.
The following chart shows the forward citations for Leap Motion
19. Page 19 of 24
© 2017 Gridlogics. All Rights Reserved.
The chart below shows the backward citations for Leap Motion
How we did it?
For citation analysis, first a group was created for all the records of Leap Motion. The created group was
selected and analyzed using the Citation Analysis option.
20. Page 20 of 24
© 2017 Gridlogics. All Rights Reserved.
KEY COMPANY ANALYSIS
The key companies considered for analysis are:
1. Leap Motion
2. Amazon
3. Oblong
4. Pixart
5. Nokia
INNOVATION TIMELINE
The chart below shows the innovation timeline representing patenting activity for hand
gesture recognition
Leap Motion had its first filing in the year 2014 and has shown consistent growth in the
subsequent years with maximum filings in 2015
How we did it?
Once the patents were added to PatSeer Projects, key assignees were filtered from the apply filters option
and innovation time line was generated within the Quick Stats Tab.
21. Page 21 of 24
© 2017 Gridlogics. All Rights Reserved.
INNOVATIONS ACROSS DIFFERENT SEGMENTS OF
GESTURE RECOGNITION
The chart below shows number of patents held by key companies across various
technology segments and applications of hand gesture recognition
How we did it?
First key companies in Hand Gesture Recognition were identified by online review and these were
analyzed with respect to the different technology segmentations for gesture recognition. The resulting
matrix was the converted to a bar chart and the chart was exported as an image.
22. Page 22 of 24
© 2017 Gridlogics. All Rights Reserved.
SUMMARY OF FILINGS BY KEY COMPANIES
The chart below shows summary report for key companies with respect to the tech
domain, Publication country and the CPC classes they are filing in
How we did it?
First key companies in hand gesture recognition were identified by online review and these were filtered
from the Apply Filter option present. Once this was done then Summary Report option was selected.
Then under “generate report for” option companies (under custom fields) option were selected.
Parameters required for analysis were selected from the option “Patent Fields” and the resulting chart
was generated and then exported as an image.
23. Page 23 of 24
© 2017 Gridlogics. All Rights Reserved.
SUMMARY
This report analyzes research trends of hand gesture recognition with a focus on Types of
Gestures, Input Devices, Sensors, Algorithms and Applications and also highlights the key
companies involved in this space.
Gesture is a symbol of physical behavior or emotional expression. It includes body gesture and
hand gesture. It falls into two categories namely static gesture and dynamic gesture. For the
former, the posture of the body or the gesture of the hand denotes a sign. For the latter, the
movement of the body or the hand conveys some messages. Gesture can be used as a tool of
communication between computer and human. It is greatly different from the traditional
hardware based methods and can accomplish human-computer interaction through gesture
recognition. Gesture recognition determines the user intent through the recognition of the
gesture or movement of the body or body parts. In the past decades, many researchers have
strived to improve the hand gesture recognition technology
Gestures involving hand act as an interface between a computer/system and the users. It works
in a way that Instead of typing with keys or tapping on a touch screen, a motion sensor
perceives and interprets movements as the primary source of data input. This is what happens
between the time a gesture is made and the computer reacts. A camera feeds image data into a
sensing device that is connected to a computer.
Kinect is Microsoft's motion sensor add-on for the Xbox 360 gaming console. The device
provides a natural user interface (NUI) that allows users to interact intuitively and without any
intermediary device, such as a controller.
The evolution of gesture technology can be useful to physically challenge and can improve
immersive gaming technology.
To get this project shared with you via PatSeer’s Patent Dashlet and to obtain access of dataset
behind this report, Contact us here:
Source: https://giphy.com/gifs/
Double click on the image to see hand gesture recognition technology example live in action.
24. Page 24 of 24
© 2017 Gridlogics. All Rights Reserved.
ABOUT PATSEER PRO
PatSeer Pro is a powerful and flexible web-based Patent Search, Analytics and Landscaping
Platform. It includes a comprehensive set of analytical tools needed for precise technology
landscaping and competitive intelligence projects such as Assignee cleanup tools, Co-occurrence
Matrices, configurable Chart Dashboards, multi-generation Citation Analysis, Text Clustering and
VizMAP – a spatial exploration and
contour mapping visualization engine.
PatSeer Pro is developed by
Gridlogics, a leading provider of
products and custom software
solutions for patent research,
management, data analysis and
project management.
For more information:
Visit us at: www.patseer.com/pro-
edition/
Create and configure
dashboards as per your
business charting
requirements
A large set of powerful
visualizations
Share dashboards across
the organization
Render multiple
visualizations in a single
view
Add interactive filters to
your dashboard
Collaborative decision
making experience via
sharable dashboards
Advanced 2D spatial
visualizations for
semantic exploration
Fully HTML5 compliant
SaaS that works across all
browsers
Powerful Customization
Options
Normalize Assignees as per
your business context
Competitor Trend Analysis
Natural Language /
Similarity Searching
Out-licensing Research
beyond citations
Powerful web based co-
occurrence matrix with
integrated real-time filters
Generate heat-maps on the
fly
A variety of Matrix Export
options
Leverage custom data
points (Custom Fields,
Categories) in Matrix
Automated Cleanup and
Grouping using Fuzzy,
Regex, Thesaurus Matching
Clustering / Topic
Identification
One Click Summary Reports
Multi-generation Citation Analyzer
Thesaurus Creation/Edit and
taxonomy management
Analyze networks of relationships
between Companies, inventors
and technologies using VizMAP
Contour Maps that highlights the
peaks and troughs of the
technology landscape
Co-Citation and Self/Non-Self
Citation Analysis
Map the landscape around your
hierarchical categories (buckets)
Powerful slide-dice options giving
full-flexibility to the analyst
Fast response times even for large
set of records
Multi-generation Citation Analyzer
Thesaurus Creation/Edit and
taxonomy management
One-stop platform for
Technology Mining, White
Space Analysis,
Tech/Economic Forecasting
and Competitive Intelligence
Powerful and flexible analytics
tools required for precision
analysis
Seamless collaboration and
workflow capabilities allow for
creating and sharing
dashboards and analytical
outputs
Integrated Global Patent
Content of PatSeer Database
PatSeer Pro
PATSEER PRO CAPABILITIES AT-A-GLANCE…