Difference between revisions of "EE586L/Projects 2013"

From WikiBiron
(Interactive Projector)
 
(26 intermediate revisions by the same user not shown)
Line 5: Line 5:
 
'''Abstract:''' ''Our project is to design and implement a drum tab reader. When we put a drum score sheet in front of the camera, the DSP board can identify what it reads and then play the rhythm.''
 
'''Abstract:''' ''Our project is to design and implement a drum tab reader. When we put a drum score sheet in front of the camera, the DSP board can identify what it reads and then play the rhythm.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/DrumRecog.pdf Poster]
  
'''Video:''' TBD
+
'''Video:''' [http://www.youtube.com/watch?v=yVGMoPqX2d0 YouTube Video]
  
 
== Interactive Projector ==
 
== Interactive Projector ==
Line 15: Line 15:
 
'''Abstract:''' ''Utilizing computer vision, Interactive Projector can convert any flat surface to a tangible screen so that people can interact with computer more freely and work more intuitively.''
 
'''Abstract:''' ''Utilizing computer vision, Interactive Projector can convert any flat surface to a tangible screen so that people can interact with computer more freely and work more intuitively.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/interactiveProjector.pdf Poster]
 +
 
 +
'''Video:''' [http://www.youtube.com/watch?v=lDrw6L-flxY YouTube Video]
 +
 
 +
== Mirrors ==
  
'''Video:''' TBD
+
'''Authors:''' [mailto:shekkizh@usc.edu, Sarath Shekkizhar], [mailto:ksavci@usc.edu Kubilay Savci], [mailto:zhan868@usc.edu Yi Zhang]
 +
 
 +
'''Abstract:''' ''A real time system that "mirrors" a users face expression on an image model has been implemented. The interaction with system is available for the user in different ways. ''
 +
 
 +
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/Face_Warping.pdf Poster]
 +
 
 +
'''Video:''' [http://www.youtube.com/watch?v=hB9ezrvGu7k YouTube Video]
  
 
== Sound Suite ==
 
== Sound Suite ==
  
'''Authors:''' [mailto:mredelma@usc.edu, Matthew Edelman], [mailto:tyronetc@usc.edu Tyrone Chen]
+
'''Authors:''' [mailto:mredelma@usc.edu Matthew Edelman], [mailto:tyronetc@usc.edu Tyrone Chen]
  
 
'''Abstract:''' ''Sound Suite is a collection of digital audio effects that can be applied to any real-time input.  The suite supports real-time layering and switching between several effects, including but not limited to distortion, wah, tremolo, delay, flanger, chorus, and reverb. The back-end audio framework is extremely flexible, supporting the development and implementation of time and frequency based effects simultaneously.''
 
'''Abstract:''' ''Sound Suite is a collection of digital audio effects that can be applied to any real-time input.  The suite supports real-time layering and switching between several effects, including but not limited to distortion, wah, tremolo, delay, flanger, chorus, and reverb. The back-end audio framework is extremely flexible, supporting the development and implementation of time and frequency based effects simultaneously.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/SoundSuite.pdf Poster]
  
'''Video:''' TBD
+
'''Video:''' [http://www.youtube.com/watch?v=Yy1L8hlm4Qk YouTube Video]
  
 
== Super Resolution Magnifier ==
 
== Super Resolution Magnifier ==
Line 35: Line 45:
 
'''Abstract:''' ''We are always expecting the clear and good-viewing images. However in real world, images are often blurring, with noise and low resolution. In this project, we want to generate a clear magnified image with super resolution technique. Many techniques can be applied and one is called super resolution.''
 
'''Abstract:''' ''We are always expecting the clear and good-viewing images. However in real world, images are often blurring, with noise and low resolution. In this project, we want to generate a clear magnified image with super resolution technique. Many techniques can be applied and one is called super resolution.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/SP.pdf Poster]
  
 
'''Video:''' TBD
 
'''Video:''' TBD
Line 45: Line 55:
 
'''Abstract:''' ''The objective of the project is to detect Facial Expressions like Smile, Surprise and Neutral Face. This is done using a combination of Face tracking algorithm and detection of change in facial features like mouth, eyes and eyebrow.''
 
'''Abstract:''' ''The objective of the project is to detect Facial Expressions like Smile, Surprise and Neutral Face. This is done using a combination of Face tracking algorithm and detection of change in facial features like mouth, eyes and eyebrow.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/Cartoonist.pdf Poster]
  
'''Video:''' TBD
+
'''Video:''' [http://www.youtube.com/watch?v=ia4vlnEelR8 YouTube Video]
  
 
== Virtual Keyboard ==
 
== Virtual Keyboard ==
Line 55: Line 65:
 
'''Abstract:''' ''The Virtual Keyboard provides an interface for users to generate keystrokes that doesn't require a set of physical switches. Using finger tracking, color thresholding and foreground segmentation, it is able to reliably track the user's keystroke gestures and produce a string of characters.''
 
'''Abstract:''' ''The Virtual Keyboard provides an interface for users to generate keystrokes that doesn't require a set of physical switches. Using finger tracking, color thresholding and foreground segmentation, it is able to reliably track the user's keystroke gestures and produce a string of characters.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/CVK.pdf Poster]
 +
 
 +
== Video Resizer ==
  
'''Video:''' TBD
+
'''Authors:''' [mailto:anujvarm@usc.edu Varma], [mailto:divyashr@usc.edu Divyashree Rao]
 +
 
 +
'''Abstract:''' ''We live in times where the same video is required to stream on different devices with different aspect ratios at very high speeds. Our project aims at retargeting video by implementing a version of seam carving improved for the DSK so as to maintain the proportions of the objects in the image while preserving the salient foreground content.
 +
 
 +
'''Video:''' [http://www.youtube.com/watch?v=UeDRBAroQUM&list=UUU15kVnJ18ubhdsEd0Nu5Bw&index=1 YouTube Video]
  
 
== Virtual Sunglasses Rendering ==
 
== Virtual Sunglasses Rendering ==
Line 65: Line 81:
 
'''Abstract:''' ''Automatic tracking of four feature points on the face. And we render a sunglasses based on those four points' information. The user can move the head around in a reasonable speed and the sunglasses still fit to the eyes.''
 
'''Abstract:''' ''Automatic tracking of four feature points on the face. And we render a sunglasses based on those four points' information. The user can move the head around in a reasonable speed and the sunglasses still fit to the eyes.''
  
'''Poster:''' TBD
+
'''Poster:''' [http://biron.usc.edu/~sungwonl/EE586/Submission_2013/VirtualSunglassesRender.pdf Poster]
  
'''Video:''' TBD
+
'''Video:''' [http://www.youtube.com/watch?v=TrDHSLKyXLo YouTube Video]

Latest revision as of 17:39, 19 May 2013

Drum Tab Reader

Authors: Chien-Tien Wu, Donglei Chen, Pei-ching Hsiao

Abstract: Our project is to design and implement a drum tab reader. When we put a drum score sheet in front of the camera, the DSP board can identify what it reads and then play the rhythm.

Poster: Poster

Video: YouTube Video

Interactive Projector

Authors: Xiyu Wang, Pengfei Wu, Biao Yang

Abstract: Utilizing computer vision, Interactive Projector can convert any flat surface to a tangible screen so that people can interact with computer more freely and work more intuitively.

Poster: Poster

Video: YouTube Video

Mirrors

Authors: Sarath Shekkizhar, Kubilay Savci, Yi Zhang

Abstract: A real time system that "mirrors" a users face expression on an image model has been implemented. The interaction with system is available for the user in different ways.

Poster: Poster

Video: YouTube Video

Sound Suite

Authors: Matthew Edelman, Tyrone Chen

Abstract: Sound Suite is a collection of digital audio effects that can be applied to any real-time input. The suite supports real-time layering and switching between several effects, including but not limited to distortion, wah, tremolo, delay, flanger, chorus, and reverb. The back-end audio framework is extremely flexible, supporting the development and implementation of time and frequency based effects simultaneously.

Poster: Poster

Video: YouTube Video

Super Resolution Magnifier

Authors: Yuzhuo Ren, Weihao Gan, Yiqian Shi, Yuchen Xie

Abstract: We are always expecting the clear and good-viewing images. However in real world, images are often blurring, with noise and low resolution. In this project, we want to generate a clear magnified image with super resolution technique. Many techniques can be applied and one is called super resolution.

Poster: Poster

Video: TBD

The Cartoonist

Authors: Anish Patil, Easswar Balasubramaniam, Mythri Thippareddy

Abstract: The objective of the project is to detect Facial Expressions like Smile, Surprise and Neutral Face. This is done using a combination of Face tracking algorithm and detection of change in facial features like mouth, eyes and eyebrow.

Poster: Poster

Video: YouTube Video

Virtual Keyboard

Authors: Alexander Nobles, Harshit Mehta

Abstract: The Virtual Keyboard provides an interface for users to generate keystrokes that doesn't require a set of physical switches. Using finger tracking, color thresholding and foreground segmentation, it is able to reliably track the user's keystroke gestures and produce a string of characters.

Poster: Poster

Video Resizer

Authors: Varma, Divyashree Rao

Abstract: We live in times where the same video is required to stream on different devices with different aspect ratios at very high speeds. Our project aims at retargeting video by implementing a version of seam carving improved for the DSK so as to maintain the proportions of the objects in the image while preserving the salient foreground content.

Video: YouTube Video

Virtual Sunglasses Rendering

Authors: Xi Guo, Jingwen Zhu, Lingchao Meng, Yi Zhang

Abstract: Automatic tracking of four feature points on the face. And we render a sunglasses based on those four points' information. The user can move the head around in a reasonable speed and the sunglasses still fit to the eyes.

Poster: Poster

Video: YouTube Video