Computer Vision Face Tracking For Use in a Perceptual User Interface

By Gary R. Bradski. 1998

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Multi-Fidelity Automatic Hyper-Parameter Tuning via Transfer Series Expansion (2019, background): “During the past decades, machine learning has achieved great successes in many applications, e.g., computer vision (Bradski 1998), recommender system (Broder 2008), ﬁnancial market analysis (Ball 2013), and so on.”
Facial feature detection using Haar classifiers (2006, background): “These algorithms suffer from the same problem; they are computationally expensive [2].”
Motion segmentation and pose recognition with motion history gradients (2000, methodology): “Possible methods of silhouette generation include stereo disparity or stereo depth subtraction [3], infrared back-lighting [12], frame differencing [13], color histogram back-projection [6], texture blob segmentation, range imagery foreground segmentation etc.”
Efficient mean-shift tracking via a new similarity measure (2005, methodology): “mean shift trackers must resort to using separate computational mechanisms such as second order moment information [5] or scale space approaches [8] to recover the scale and other information of the targets.”
Real-Time Tracking via On-line Boosting (2006, background): “The efficient and robust tracking of objects in complex environments is important for a variety of applications including video surveillance [25], autonomous driving [1] or human-computer interaction [5, 4].”
Fast multiple object tracking via a hierarchical particle filter (2005, methodology): “Model-based tracking algorithms incorporate a priori information about the objects to develop representations such as skin complexion, body blobs, kinematic skeleton, silhouettes or layer information [38, 36, 5, 33, 32, 6].”
Tracking Using CamShift Algorithm and Multiple Quantized Feature Spaces (2004, methodology): “The histogram used in Bradski (1998) consists of the hue channel in HSV color space, however multidimensional histograms from any color space may be used.”
A brief introduction to OpenCV (2012, methodology): “In addition, a more advanced version where the size and the window orientation can change is implemented as well under the name CamShift algorithm [7].”
Towards reliable multimodal sensing in aware environments (2001, methodology): “The camshift algorithm is used to perform occupant tracking with both cameras [3].”
Mean-shift blob tracking through scale space (2003, methodology): “[1] uses moments of the sample weight image to compute blob scale and orientation.”
On-line selection of discriminative tracking features (2003, methodology): “In this ideal case, tracking could be achie ved simply by thresholding at zero and computing the object center and rough shape using the met hod of moments (see also [4]).”