Papers

A Real Time Hand Tracking System for Interactive Applications

In this paper the hand tracking is perfomed with color detection, with this technique we can segment the hand for de background. The tracking is in real time and is robust under differents lighting conditions. This technique can be applied in video games, virtual mouse, aumented reality etc.

A Robust Method for Hand TrackingUsing Mean-shift Algorithm and Kalman Filter in Stereo Color Image Sequences

The hand tracking is performed through depth map 3D, Mean – shift for hand segmentation, and the kalman filter for estímate the hand position. This technique allow a partial overlapping of the hands and face.

An Adaptive Self-Organizing Color Segmentation Algorithm with Application to Robust Real-time Human HandLocalization

This use the color segmentation for the hand tracking, they do this through self-organizing map and others techniques that i don’t understand.

ANÁLISIS DE IMÁGENES: SISTEMA DE SEGUIMIENTO DE PERSONAS

Se realiza el seguimiento de personas a través de una cámara de seguridad, mediante la cual se identifican grupos de personas y se realiza un seguimiento individual unas vez se hanseparado, el algoritmo Mean-Shift permite un mayor rendimiento de este al momento de oclusiones.

* Efficient Model-based 3D Tracking of Hand Articulations using Kinect

In this work the, They performed the hand tracking and the articulations using the kinect sensor. Through the data of the sensor they segment the hand and they do the tracking of the hand in 3D, just with one camera. Thetracking is almost in real time and the computational cost are low but they use the GPU of kinect sensor for the processing.

* EVALUATION OF REAL-TIME HAND MOTION TRACKING USING A RANGE CAMERA AND THE MEAN-SHIFT ALGORITHM

In this paper they work with a range camera (sr4000), this camera have infrareds. They want test the efectivity of hand tracking using the Mean- Shift algorithm, thisalgorithm usually is used with the Kalman Filter for more performance.The accuary is good and is fast enough for work in real time.

Hand Tracking via Efficient Database Indexing

Head and Hands 3D Tracking in Real Time by the EM algorithm

This work use the EM (Expectation Maximisation), this algorithm work with likehood of pixels, is like a stimate and iterations of pixels. In the method usein this paper don´t use segmentation, only skin color detection and EM algorithm, the program work near to real time (12 Hz) and detected the hands and the head of person. Work with 2 cameras for detected of position. The relevant information used is the position and orientation of the objetives.(ellipsoids).

MMX-Accelerated Real-Time Hand Tracking System.

The hand tracking is performed witha cheap camera, personal computer with no specialized image processing hardware, and the MMX (Multi-Media Instruction set ) of Intel Pentium chip family. The algorithm used was the CamShift together with Kalmar Filter. The system work at 30 fps, is robust and have low costs of hadware. Work under Windows 2000.

REAL TIME HAND TRACKING AND 3D GESTURE RECOGNITION FOR INTERACTIVE INTERFACESUSING HMM

They work with 2 cameras for 3D recognition, the Kalman filter in 2D and 3D for noise reduction and the HMM ( Hidden Markov Models ) for gesture recognition. Used a object for the calibration system and the recognition be more easy. First we have applied several image processing algorithms to detect markers in bitmap images acquired from simple web cameras. Then we have applied 3Dreconstruction from stereo images. This paper work with gestures recognition for interact with PC applications.

* Real –Time Hand Tracking and Gesture Recognition for Human-Computer Interaction

In this paper the objective is make a videogame controller with hand gestures. The hand recognition have three steps: hand segmentation, hand tracking and gesture recognition. The segmentation is...