EE586L/Projects 2011
Contents
Alter Ego
Authors: Nirmal Patel, Vipresh Gangwal
Abstract: Alter Ego; latin "the other I" here refers to one's own reflection. The project acts like a mirror. One can put on the trioscopic goggles and look into a 3D image of oneself in the LCD screen. This colorful 3D represntation is achived by generating Anaglyphs.
Poster:
Video:
Blu-Rabbit
Authors: Evan Lee, Jose Herrera, Ross MacKinnon
Abstract: In an attempt to create a more ergonomic and natural interface between humans and computers, this project strives to utilize common communication techniques to replace the hardware used today. The outdated and unnatural mouse is not the optimal way to create a comfortable experience for the user, and thus this project utilizes intuitive gestures to achieve the same end. By developing and tweaking algorithms for finger tracking and gesture recognition, the members of the Blu-Rabbit team show how interaction between humans and computers may mimic interactions seen between humans.
Poster:
Video:
Cam Hockey
Authors:Anoop Panicker, Mihir Thakkar, Sanjay Jain, Vinit Adkar
Abstract: Cam Hockey is a video camera-based interactive version of the popular game Air Hockey. The goal of the game is to control your player on the screen using your hand movements (with a color marker) and hit the ball towards the opposite side of the screen controlled by another player.If a player is unable to hit the ball, the other player scores a goal. The objective of the game is to score as many goals as you can while outscoring your opponent.
Poster:
Video:
CamMouse
Authors: Yen-Cheng Chen, Chan-Yi Ou, Yu-Ting Wu, Jin-Wei Lee
Abstract: Our goal is develop a virtual computer cursor control interface which can control computer cursor and do right click and left click as normal computer mouse. This interface will be implemented by video camera. Users with this system can control the computer cursor with bare hands (or with some color markers stick on hand). This idea is inspired by our daily life; although touch panel is getting better and better nowadays, using mouse is still the easiest and most effort-saving way to control computer cursor. However, to use computer mouse, a flat and unreflected surface is needed. By using this system, we don’t need to bring a real mouse and can be used in any kind of surface.
Poster:
Video:
Dancing Machine
Authors: Yue Pan, Chenchen Wang
Abstract: Dancing Machine Modeling: 1) Judge the behavior of player wearing color markers based on color detection algorithm. If the player locates the right colored marker into the requested space area at correct moment, one score will be added. 2) The speed of the game(flying arrow speed)is in pace with the chosen real-time music, based on beat detection algorithm. 3) The interface is divided into three parts: arrow panel, score showing panel and real-time camera video panel. 4) The game complexity could be changed by switches of the platform.
Poster:
Video:
Fatigue Driving Warning System
Authors: Young Chun Ahn, Lu Liu, Yinxiao Zhang
Abstract: A system that monitors the user with a camera and send out warnings when it's detected that the user is dozing off.
Poster:
Video:
Go Five
Authors: Fei Yu, Weijie Zhao, Zhe Wang
Abstract: Using motion based control techniques in games is becoming gradually more popular nowadays. Mobile cameras and real-time video processing take place of traditional input devices, which may provide much better game experience with much more freedom and joy. In this project, gestures of fingers are used as the control factor. Skin color extraction and pattern matching are adopted to detect the position of finger tips to achieve low complexity for real-time interactivity. In addition to that, modified low pass filter is applied to reduce the influence of noise and therefore keep the robustness of the detection. More than five kinds of control commands defined by distinct finger gestures are designed user-friendly. Overall, by introducing this novel way to a real-time interactive game, players are able to enjoy fancy gaming experiences by simply moving their fingers in front of a camera.
Poster:
Video:
Got Ya
Authors: Yu-Chun Yang, Chun-Kang Chen, Yung-Chun Chen
Abstract: We are building a surveillance system to track and label the people in front of the camera. Once people step inside the monitor area, the system will show the outline of people and each people will be assigned a bounding box with different colors from others. The goal of our system is to accurately count the number of people in the monitor area.
Poster:
Video:
hEEEro
Authors: Xiang Fu, Lidou Wang, Yingbu Kou
Abstract: Our project helps to establish a renewable background (BG) model using Gaussian Mixture Models (GMM) algorithm based on frame sequences, and recognize the foreground (FG) region in real time, followed by adding pre-stored image or video background. The main applications involve movie industry, weather forecast, photograph with super stars, surveillance, background modeling, etc.
Poster:
Video:
Match & OCR
Authors: Chun-Ting Huang, Kuan-Ming Lin, Yen-Feng Lee
Abstract: Our project includes one mini game and Chinese number OCR system. Firstly, the mini game utilizes marker tracking to check the marker matches the position on the screen, and if it matches, the score also increases. Secondly, we build writing interface for input in the Chinese number OCR system, and the system can recognize the input information as Chinese number or unknown.
Poster:
Video:
Midas Hands
Authors: Geun Lee, Suk Jin Lee, Shafil ahmed, Christian Idyllel
Abstract: Depth data from the Kinect is used for hand gesture recognition to manipulate on-screen images through geometric modifications. These modifications include scaling and translation.
Poster:
Video:
Pitchin' Wavelets
Authors: Balamurali Govindaraju, Yule Wu
Abstract: Pitch Shifting by Wavelets.
Poster:
Video:
Project Conductor
Authors: Nitin Balajee Ravi, Prasanna Dhamodaran, Sung Kwang Cho
Abstract: The objective of the project is to change the tempo of music real-time using gestures. We also have multiple/single instruments playing some music whose tempo is varied.
Poster:
Video:
Super Advertisement
Authors: Jian Chen, Boyang Li, Chen Ling
Abstract: A method for automatic insertion of image into a real-time video sequence is implemented. The algorithm is based on Harris corner detection and image mapping.
Poster:
Video:
Tiro
Authors: Rashmi Deshmukh, Hardik Shah
Abstract: For our project we got inspired by the progress in the field of HCI (Human – Computer Interaction) and its multidisciplinary nature. The development in human gesture recognition and how well the computer understands the body motion. We are implementing a small game by using Monte Carlo tracking technique to track the marker on the wrist of a human arm. The game involves throwing a block in one of the slots shown on the display. The player first selects the slot where he would throw the block after which he is expected to throw with the same speed in the same slot 3 times to win the game. He would be given 10 chances to do. The main challenge of our project was to increase accuracy of tracking the marker and increase repeatability of the result after a throwing action is performed by the user standing at any distance from the camera. Since the game involves fast motion tracking of the arm, we had to give special attention to the speed factor keeping in mind the drawbacks of the DSP kit.
Poster:
Video:
