Showing papers by "Henry Allan Rowley published in 1997"

Method for finding all frontal faces in arbitrarily complex visual scenes

Abstract: A method for detecting a face in an image includes the steps of applying the image to a first classification tool that determines a rotational angle for rotating the image that makes the image most resemble an upright face; rotating the image by the rotational angle determined by the first classification tool; and applying the rotated image to a second classification tool, which determines whether the rotated image represents a frontal face or not.

Method for detecting human body motion in frames of a video sequence

Abstract: In a computerized method, a moving object is detected in a sequence of frames of a video of a scene. Each of the frames includes a plurality of pixels representing measured light intensity values at specific locations in the scene. The pixels are organized in a regularized pattern in a memory. The object is modeled as a branched kinematic chain composed of links connected at joints. The frames are iteratively segmented by assigning groups of pixels having like pixel motion to individual links, while estimating motion parameters for the groups of pixels assigned to the individual links until the segmented pixels and their motion parameters converge and can be identified with the moving object as modeled by the kinematic chain.

Analyzing articulated motion using expectation-maximization

Abstract: We present a novel application of the Expectation-Maximization algorithm to the global analysis of articulated motion. The approach utilizes a kinematic model to constrain the motion estimates, producing a segmentation of the flow field into parts with different articulated motions. Experiments with synthetic and real images are described.

Integrating Text and Face Detection for Finding Informative Poster Frames

Abstract: Digital video is rapidly becoming an important source of information and entertainment, and is used in a host of multimedia applications. With the size of digital video collections growing to many thousands of hours, technology is needed to allow rapid browsing of videos. One way to summarize a video is to select poster frames to represent segments of the video. Previous techniques for extracting poster frames were based on scene segmentation, using color histograms or optical flow. To provide more informative poster frames, this work combines algorithms for extracting image content, specifically faces and on-screen text, with existing scene segmentation technology.