SlideShare a Scribd company logo
1 of 1
Pratheba Selvaraju
( 4 Homework + 1 Project )
(University of Washington)
Real-Time Designing and Modeling
Conclusions
Idea:
Introduction
Real Time Designing and Modelling:
Everyday engagement in
Conversation over video chat
• It will be useful to present
a design idea on the spot without
Use of any special software.
• Fun to play games where you
Model your own character.
• Show the person in other end
of video conference a new
Idea for a car design.
Approach- Phase I
Probability Mask and Bounding Box:
• Skin-Color-Probability mask
•
•
•
~ 20 frames
• Optical-flow-Probability mask
•
• Region of Interest
Centroid and BoundingBox
Results :
•Due to high number of processing,
frame rate is low, which results in
not a very smooth drawing.
•Very high speed movements
could not be tracked.
• To design withoutadditional software
• No additional hardware
• Must work with lowresolution camera
• Must be real-time withlittle to no-preprocessing
• Use of Hand / FingerTip to draw
and design.
• Use of motion to detect which object
To track for drawing
• Get the drawing to create 3D model
• Augment it to the video
Requirement
System Overview
Use case
Meeting Presentation
Gaming Application
• Face is detected from each frame
• ROI of face region is averaged
• HSV range of skin color is obtained, (bin
size = range)
• Hue Range is Obtained from HSV
.
.
.
Averaged face
HSV range
Haar Cascade classifier
Viola-Jones
(Frontal face)
Hue Mask
References
•Araki, R. ; Waseda Univ., Tokyo, Japan ;
Gohshi, S. ; Ikenaga, T. Real-time both hands
tracking using CAMshift with motion mask and
probability reduction by motion prediction,
Signal & Information Processing Association
Annual Summit and Conference (APSIPA ASC),
2012 Asia-Pacific
•Dorin Comaniciu,, Mean Shift : A Robust
approach towards feature space, IEEE
transactions on Pattern Analysis and Machine
intelligence
•Jean-yves Bouguet, Pyramidal implementation
of the Lucas Kanade feature tracker, Intel
Corporation, Microprocessor Research Labs,
2000.
Previous
Frame
Current
Frame
Grid
Features
Where is Probability mask
and is the x and y flow
vector
Pyramidal Lucas Kanade
Dense Optical
Flow
FlowmaskDense Optical
Flow
Normalize to range
between 0 – 255
= threshold = 255
= threshold = 0
• Optical flow image is smoothed, filtered and dilated
(~ 3 iteration) to remove noise.
• Extract image from FlowMask with range 200-255
• Normalize the extracted image to 0-255
• Obtain Contours and convex hull
Three coin algorithm
• Ability to draw in real-time with
Only webcam (low resolution).
• Can be used to track any moving
With better accuracy than traditional
Camshift.
ContourImage Convex Hull
FlowmaskHue Mask
• Regular camshift uses only the color
probability, which might result in incorrect
tracking due to noise or object with same
color range
• With addition of flow mask, it correctly
identifies the object to be tracked.
• Depending upon speed of object,
threshold level is adjusted.
Bounding box andcentroid used for initial
setting initial region ofinterest for tracking
where, 0 ≤ α ≤ 1 and 0 ≤ β ≤ 1
• When object start moving, camshaft tracks the
object
• Previous and current tracking box positions are
obtained.
• Line is drawn between center of previous and
current box position
Future Tasks:
•Improve performance processing.
•Convert 2D drawing into 3D model
real-time
(either via classification of shapes
or by rendering exact 3D shape of
2D drawing)
•Augment the 3D model in video in
real-time
•Send it via network to be viewed
by other person in chat

More Related Content

What's hot

A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...
A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...
A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...Electronic Arts / DICE
 
Advanced Lighting for Interactive Applications
Advanced Lighting for Interactive ApplicationsAdvanced Lighting for Interactive Applications
Advanced Lighting for Interactive Applicationsstefan_b
 
Screen space reflections on Epsilon Engine
Screen space reflections on Epsilon EngineScreen space reflections on Epsilon Engine
Screen space reflections on Epsilon EngineImanol Fotia
 
Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...
Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...
Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...Lviv Startup Club
 
「原神」におけるコンソールプラットフォーム開発
「原神」におけるコンソールプラットフォーム開発「原神」におけるコンソールプラットフォーム開発
「原神」におけるコンソールプラットフォーム開発Unity Technologies Japan K.K.
 
Authoring of procedural rocks in The Blacksmith realtime short
Authoring of procedural rocks in The Blacksmith realtime shortAuthoring of procedural rocks in The Blacksmith realtime short
Authoring of procedural rocks in The Blacksmith realtime shortVesselin Efremov
 
Render techniques overview
Render techniques overviewRender techniques overview
Render techniques overviewVostrikov Arkady
 
Shiny Pixels and Beyond: Real-Time Raytracing at SEED
Shiny Pixels and Beyond: Real-Time Raytracing at SEEDShiny Pixels and Beyond: Real-Time Raytracing at SEED
Shiny Pixels and Beyond: Real-Time Raytracing at SEEDElectronic Arts / DICE
 
Nirvigna - Rendering Hi-Res graphics on commodity cluster
Nirvigna - Rendering Hi-Res graphics on commodity clusterNirvigna - Rendering Hi-Res graphics on commodity cluster
Nirvigna - Rendering Hi-Res graphics on commodity clusterAbhishek Parolkar
 
Visibility Optimization for Games
Visibility Optimization for GamesVisibility Optimization for Games
Visibility Optimization for GamesUmbra
 
The Rendering Technology of Killzone 2
The Rendering Technology of Killzone 2The Rendering Technology of Killzone 2
The Rendering Technology of Killzone 2Guerrilla
 

What's hot (12)

A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...
A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...
A Certain Slant of Light - Past, Present and Future Challenges of Global Illu...
 
Advanced Lighting for Interactive Applications
Advanced Lighting for Interactive ApplicationsAdvanced Lighting for Interactive Applications
Advanced Lighting for Interactive Applications
 
Screen space reflections on Epsilon Engine
Screen space reflections on Epsilon EngineScreen space reflections on Epsilon Engine
Screen space reflections on Epsilon Engine
 
Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...
Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...
Данило Ульянич “C89 OpenGL for ARM microcontrollers on Cortex-M. Basic functi...
 
「原神」におけるコンソールプラットフォーム開発
「原神」におけるコンソールプラットフォーム開発「原神」におけるコンソールプラットフォーム開発
「原神」におけるコンソールプラットフォーム開発
 
Authoring of procedural rocks in The Blacksmith realtime short
Authoring of procedural rocks in The Blacksmith realtime shortAuthoring of procedural rocks in The Blacksmith realtime short
Authoring of procedural rocks in The Blacksmith realtime short
 
Render techniques overview
Render techniques overviewRender techniques overview
Render techniques overview
 
The Unique Lighting of Mirror's Edge
The Unique Lighting of Mirror's EdgeThe Unique Lighting of Mirror's Edge
The Unique Lighting of Mirror's Edge
 
Shiny Pixels and Beyond: Real-Time Raytracing at SEED
Shiny Pixels and Beyond: Real-Time Raytracing at SEEDShiny Pixels and Beyond: Real-Time Raytracing at SEED
Shiny Pixels and Beyond: Real-Time Raytracing at SEED
 
Nirvigna - Rendering Hi-Res graphics on commodity cluster
Nirvigna - Rendering Hi-Res graphics on commodity clusterNirvigna - Rendering Hi-Res graphics on commodity cluster
Nirvigna - Rendering Hi-Res graphics on commodity cluster
 
Visibility Optimization for Games
Visibility Optimization for GamesVisibility Optimization for Games
Visibility Optimization for Games
 
The Rendering Technology of Killzone 2
The Rendering Technology of Killzone 2The Rendering Technology of Killzone 2
The Rendering Technology of Killzone 2
 

Viewers also liked

[Arabic] Introduction to Kinect Article
[Arabic] Introduction to Kinect Article[Arabic] Introduction to Kinect Article
[Arabic] Introduction to Kinect ArticleIslam Negm
 
مقياس التسارع
مقياس التسارعمقياس التسارع
مقياس التسارعnabeelasd
 
Algorithm in discovering and correcting words errors in a dictionary or any w...
Algorithm in discovering and correcting words errors in a dictionary or any w...Algorithm in discovering and correcting words errors in a dictionary or any w...
Algorithm in discovering and correcting words errors in a dictionary or any w...kinan keshkeh
 
Características de la lectura crítica
Características de la lectura críticaCaracterísticas de la lectura crítica
Características de la lectura críticampardoc
 
Rapid object detection using boosted cascade of simple features
Rapid object detection using boosted  cascade of simple featuresRapid object detection using boosted  cascade of simple features
Rapid object detection using boosted cascade of simple featuresHirantha Pradeep
 
Imagemagickknowhow
ImagemagickknowhowImagemagickknowhow
ImagemagickknowhowYo Ya
 
Costo de produccion power point 2016
Costo de produccion power point 2016Costo de produccion power point 2016
Costo de produccion power point 2016mandreinahp
 
Medios alternos teorías que contribuyen a la resolución pacífica de los confl...
Medios alternos teorías que contribuyen a la resolución pacífica de los confl...Medios alternos teorías que contribuyen a la resolución pacífica de los confl...
Medios alternos teorías que contribuyen a la resolución pacífica de los confl...miguel_alejandrogp
 
4 أنواع خوارزميات التشفير
4  أنواع خوارزميات التشفير4  أنواع خوارزميات التشفير
4 أنواع خوارزميات التشفيررياض خليفة
 
A neural ada boost based facial expression recogniton System
A neural ada boost based facial expression recogniton SystemA neural ada boost based facial expression recogniton System
A neural ada boost based facial expression recogniton SystemInternational Islamic University
 
Stop Don't R&R Contractors Flyer
Stop Don't R&R Contractors FlyerStop Don't R&R Contractors Flyer
Stop Don't R&R Contractors FlyerCraig Harrington
 
state logic progamming
state logic progamming state logic progamming
state logic progamming nabeelasd
 
100 of the most influential German Videogames | Ralf C. Adam
100 of the most influential German Videogames | Ralf C. Adam100 of the most influential German Videogames | Ralf C. Adam
100 of the most influential German Videogames | Ralf C. AdamRalf C. Adam
 
TWU_NEWS_winter_2012_web(1)
TWU_NEWS_winter_2012_web(1)TWU_NEWS_winter_2012_web(1)
TWU_NEWS_winter_2012_web(1)Iva Tora
 
Юридична математика
Юридична математикаЮридична математика
Юридична математикаzdwango
 

Viewers also liked (20)

[Arabic] Introduction to Kinect Article
[Arabic] Introduction to Kinect Article[Arabic] Introduction to Kinect Article
[Arabic] Introduction to Kinect Article
 
مقياس التسارع
مقياس التسارعمقياس التسارع
مقياس التسارع
 
Road Extractor
Road ExtractorRoad Extractor
Road Extractor
 
Algorithm in discovering and correcting words errors in a dictionary or any w...
Algorithm in discovering and correcting words errors in a dictionary or any w...Algorithm in discovering and correcting words errors in a dictionary or any w...
Algorithm in discovering and correcting words errors in a dictionary or any w...
 
CONGETUR - HBD CONSULTING ANDALUCIA
CONGETUR - HBD CONSULTING ANDALUCIACONGETUR - HBD CONSULTING ANDALUCIA
CONGETUR - HBD CONSULTING ANDALUCIA
 
Invitation kit
Invitation kitInvitation kit
Invitation kit
 
Características de la lectura crítica
Características de la lectura críticaCaracterísticas de la lectura crítica
Características de la lectura crítica
 
Rapid object detection using boosted cascade of simple features
Rapid object detection using boosted  cascade of simple featuresRapid object detection using boosted  cascade of simple features
Rapid object detection using boosted cascade of simple features
 
Imagemagickknowhow
ImagemagickknowhowImagemagickknowhow
Imagemagickknowhow
 
Costo de produccion power point 2016
Costo de produccion power point 2016Costo de produccion power point 2016
Costo de produccion power point 2016
 
Medios alternos teorías que contribuyen a la resolución pacífica de los confl...
Medios alternos teorías que contribuyen a la resolución pacífica de los confl...Medios alternos teorías que contribuyen a la resolución pacífica de los confl...
Medios alternos teorías que contribuyen a la resolución pacífica de los confl...
 
4 أنواع خوارزميات التشفير
4  أنواع خوارزميات التشفير4  أنواع خوارزميات التشفير
4 أنواع خوارزميات التشفير
 
A neural ada boost based facial expression recogniton System
A neural ada boost based facial expression recogniton SystemA neural ada boost based facial expression recogniton System
A neural ada boost based facial expression recogniton System
 
Stop Don't R&R Contractors Flyer
Stop Don't R&R Contractors FlyerStop Don't R&R Contractors Flyer
Stop Don't R&R Contractors Flyer
 
state logic progamming
state logic progamming state logic progamming
state logic progamming
 
MESH DECOR-PRESENTATION 1
MESH DECOR-PRESENTATION 1MESH DECOR-PRESENTATION 1
MESH DECOR-PRESENTATION 1
 
100 of the most influential German Videogames | Ralf C. Adam
100 of the most influential German Videogames | Ralf C. Adam100 of the most influential German Videogames | Ralf C. Adam
100 of the most influential German Videogames | Ralf C. Adam
 
TWU_NEWS_winter_2012_web(1)
TWU_NEWS_winter_2012_web(1)TWU_NEWS_winter_2012_web(1)
TWU_NEWS_winter_2012_web(1)
 
Юридична математика
Юридична математикаЮридична математика
Юридична математика
 
Berlin - January 2016
Berlin - January 2016Berlin - January 2016
Berlin - January 2016
 

Similar to FinalPoster

Sem 2 Presentation
Sem 2 PresentationSem 2 Presentation
Sem 2 PresentationShalom Cohen
 
Digital Dynamic Markers in augmented Reality
Digital Dynamic Markers in augmented RealityDigital Dynamic Markers in augmented Reality
Digital Dynamic Markers in augmented RealityAyan Sinha
 
pick and place robotic arm
pick and place robotic armpick and place robotic arm
pick and place robotic armANJANA ANILKUMAR
 
Mainprojpresentation 150617092611-lva1-app6892
Mainprojpresentation 150617092611-lva1-app6892Mainprojpresentation 150617092611-lva1-app6892
Mainprojpresentation 150617092611-lva1-app6892ANJANA ANILKUMAR
 
Overview of Computer Vision For Footwear Industry
Overview of Computer Vision For Footwear IndustryOverview of Computer Vision For Footwear Industry
Overview of Computer Vision For Footwear IndustryTanvir Moin
 
"High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro...
"High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro..."High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro...
"High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro...Edge AI and Vision Alliance
 
Cahall Final Intern Presentation
Cahall Final Intern PresentationCahall Final Intern Presentation
Cahall Final Intern PresentationDaniel Cahall
 
Introduction talk to Computer Vision
Introduction talk to Computer Vision Introduction talk to Computer Vision
Introduction talk to Computer Vision Chen Sagiv
 
cebas Visual Technology: VFX and Render software - presentation 2015
cebas Visual Technology: VFX and Render software - presentation 2015cebas Visual Technology: VFX and Render software - presentation 2015
cebas Visual Technology: VFX and Render software - presentation 2015Zedar Thokme
 
TMD2063 | Digital Animation - Chapter 3
TMD2063 | Digital Animation - Chapter 3TMD2063 | Digital Animation - Chapter 3
TMD2063 | Digital Animation - Chapter 3Diyana Harithuddin
 
支援DSL的嵌入式圖形操作環境
支援DSL的嵌入式圖形操作環境支援DSL的嵌入式圖形操作環境
支援DSL的嵌入式圖形操作環境John Chou
 
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine LearningMakine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine LearningAli Alkan
 
Computer Graphics Notes
Computer Graphics NotesComputer Graphics Notes
Computer Graphics NotesGurpreet singh
 
Real Time Object Dectection using machine learning
Real Time Object Dectection using machine learningReal Time Object Dectection using machine learning
Real Time Object Dectection using machine learningpratik pratyay
 
Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...
Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...
Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...Aritra Sarkar
 
SIGGRAPH ASIA 2012 "2x3D"
SIGGRAPH ASIA 2012 "2x3D"SIGGRAPH ASIA 2012 "2x3D"
SIGGRAPH ASIA 2012 "2x3D"Akihiko Shirai
 

Similar to FinalPoster (20)

Sem 2 Presentation
Sem 2 PresentationSem 2 Presentation
Sem 2 Presentation
 
IMAGE PROCESSING
IMAGE PROCESSINGIMAGE PROCESSING
IMAGE PROCESSING
 
Digital Dynamic Markers in augmented Reality
Digital Dynamic Markers in augmented RealityDigital Dynamic Markers in augmented Reality
Digital Dynamic Markers in augmented Reality
 
pick and place robotic arm
pick and place robotic armpick and place robotic arm
pick and place robotic arm
 
Mainprojpresentation 150617092611-lva1-app6892
Mainprojpresentation 150617092611-lva1-app6892Mainprojpresentation 150617092611-lva1-app6892
Mainprojpresentation 150617092611-lva1-app6892
 
Overview of Computer Vision For Footwear Industry
Overview of Computer Vision For Footwear IndustryOverview of Computer Vision For Footwear Industry
Overview of Computer Vision For Footwear Industry
 
"High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro...
"High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro..."High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro...
"High-resolution 3D Reconstruction on a Mobile Processor," a Presentation fro...
 
3D modeling and animation
3D modeling and animation3D modeling and animation
3D modeling and animation
 
Cahall Final Intern Presentation
Cahall Final Intern PresentationCahall Final Intern Presentation
Cahall Final Intern Presentation
 
Introduction talk to Computer Vision
Introduction talk to Computer Vision Introduction talk to Computer Vision
Introduction talk to Computer Vision
 
cebas Visual Technology: VFX and Render software - presentation 2015
cebas Visual Technology: VFX and Render software - presentation 2015cebas Visual Technology: VFX and Render software - presentation 2015
cebas Visual Technology: VFX and Render software - presentation 2015
 
Scan-to-BIM english
Scan-to-BIM englishScan-to-BIM english
Scan-to-BIM english
 
TMD2063 | Digital Animation - Chapter 3
TMD2063 | Digital Animation - Chapter 3TMD2063 | Digital Animation - Chapter 3
TMD2063 | Digital Animation - Chapter 3
 
Virtual Mouse
Virtual MouseVirtual Mouse
Virtual Mouse
 
支援DSL的嵌入式圖形操作環境
支援DSL的嵌入式圖形操作環境支援DSL的嵌入式圖形操作環境
支援DSL的嵌入式圖形操作環境
 
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine LearningMakine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
Makine Öğrenmesi ile Görüntü Tanıma | Image Recognition using Machine Learning
 
Computer Graphics Notes
Computer Graphics NotesComputer Graphics Notes
Computer Graphics Notes
 
Real Time Object Dectection using machine learning
Real Time Object Dectection using machine learningReal Time Object Dectection using machine learning
Real Time Object Dectection using machine learning
 
Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...
Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...
Computer-Vision based Centralized Multi-agent System on Matlab and Arduino Du...
 
SIGGRAPH ASIA 2012 "2x3D"
SIGGRAPH ASIA 2012 "2x3D"SIGGRAPH ASIA 2012 "2x3D"
SIGGRAPH ASIA 2012 "2x3D"
 

FinalPoster

  • 1. Pratheba Selvaraju ( 4 Homework + 1 Project ) (University of Washington) Real-Time Designing and Modeling Conclusions Idea: Introduction Real Time Designing and Modelling: Everyday engagement in Conversation over video chat • It will be useful to present a design idea on the spot without Use of any special software. • Fun to play games where you Model your own character. • Show the person in other end of video conference a new Idea for a car design. Approach- Phase I Probability Mask and Bounding Box: • Skin-Color-Probability mask • • • ~ 20 frames • Optical-flow-Probability mask • • Region of Interest Centroid and BoundingBox Results : •Due to high number of processing, frame rate is low, which results in not a very smooth drawing. •Very high speed movements could not be tracked. • To design withoutadditional software • No additional hardware • Must work with lowresolution camera • Must be real-time withlittle to no-preprocessing • Use of Hand / FingerTip to draw and design. • Use of motion to detect which object To track for drawing • Get the drawing to create 3D model • Augment it to the video Requirement System Overview Use case Meeting Presentation Gaming Application • Face is detected from each frame • ROI of face region is averaged • HSV range of skin color is obtained, (bin size = range) • Hue Range is Obtained from HSV . . . Averaged face HSV range Haar Cascade classifier Viola-Jones (Frontal face) Hue Mask References •Araki, R. ; Waseda Univ., Tokyo, Japan ; Gohshi, S. ; Ikenaga, T. Real-time both hands tracking using CAMshift with motion mask and probability reduction by motion prediction, Signal & Information Processing Association Annual Summit and Conference (APSIPA ASC), 2012 Asia-Pacific •Dorin Comaniciu,, Mean Shift : A Robust approach towards feature space, IEEE transactions on Pattern Analysis and Machine intelligence •Jean-yves Bouguet, Pyramidal implementation of the Lucas Kanade feature tracker, Intel Corporation, Microprocessor Research Labs, 2000. Previous Frame Current Frame Grid Features Where is Probability mask and is the x and y flow vector Pyramidal Lucas Kanade Dense Optical Flow FlowmaskDense Optical Flow Normalize to range between 0 – 255 = threshold = 255 = threshold = 0 • Optical flow image is smoothed, filtered and dilated (~ 3 iteration) to remove noise. • Extract image from FlowMask with range 200-255 • Normalize the extracted image to 0-255 • Obtain Contours and convex hull Three coin algorithm • Ability to draw in real-time with Only webcam (low resolution). • Can be used to track any moving With better accuracy than traditional Camshift. ContourImage Convex Hull FlowmaskHue Mask • Regular camshift uses only the color probability, which might result in incorrect tracking due to noise or object with same color range • With addition of flow mask, it correctly identifies the object to be tracked. • Depending upon speed of object, threshold level is adjusted. Bounding box andcentroid used for initial setting initial region ofinterest for tracking where, 0 ≤ α ≤ 1 and 0 ≤ β ≤ 1 • When object start moving, camshaft tracks the object • Previous and current tracking box positions are obtained. • Line is drawn between center of previous and current box position Future Tasks: •Improve performance processing. •Convert 2D drawing into 3D model real-time (either via classification of shapes or by rendering exact 3D shape of 2D drawing) •Augment the 3D model in video in real-time •Send it via network to be viewed by other person in chat