Canon Kabushiki Kaisha

Xerox Corporation

Michael James Taylor

Simon Michael Rowe

William M Newman

Jebu Jacob Rajan

Joe S C Lee

Anthony Zappala

Three-dimensional computer model data, moving image data or still image data showing at least one person is stored in an archive database, together with additional information to improve the searching and retrieval of data therefrom. The additional information includes view parameter datawhich defines at whom or what each person is looking during each predetermined period of time or image. Text datawhich comprises words associated with the person, and viewing histogram datawhich, for each period of text data defines the percentage of time that the speaking person spent looking at each other person or object, may also be stored.

Image processing apparatus and database

Image data from cameras showing movement of a number of people, and sound data, is archived and processed to determine the position and orientation of each person's head and to determine at whom each person is looking. The speaker is determined by determining at which person most people are looking. Alternatively, the sound data is processed to determine the direction from which the sound came, and it is determined who is speaking by determining which person's head is in a position corresponding to the direction from which the sound came. The personal speech recognition parameters for the speaker are selected and used to convert the sound data to text data. Image data to be archived is chosen by selecting the camera which best shows the speaker and the participant to whom he is speaking. Data is stored in a meeting archive database.

Processing apparatus for determining which person in a group is speaking

Image data from a plurality of cameras---showing the movements of a number of people, for example in a meeting, and sound data from a directional microphone arrayis processed by a computer processing apparatusto archive the data in a meeting archive database. The image data is processed to determine the three-dimensional position and orientation of each person's head and to determine at whom each person is looking. The sound data is processed to determine the direction from which the sound came. Processing is carried out to determine who is speaking by determining which person has his head in a position corresponding to the direction from which the sound came. Having determined which person is speaking, the personal speech recognition parameters for that person are selected and used to convert the sound data to text data. Image data to be archived is chosen by selecting the camera which best shows the speaking participant and the participant to whom he is speaking. Image data, sound data, text data and data defining at whom each person is looking is stored in the meeting archive database

Image processing apparatus

A face-up document scanning apparatus stitches views from multiple video cameras together to form a composite image. The document scanning apparatus includes an image acquisition system and a frame merger module. The image acquisition, which is mounted over the surface of a desk on which a hardcopy document is placed, has two video cameras for simultaneously recording two overlapping images of different portions of the hardcopy document. By overlapping a portion of the recorded images, the document scanning apparatus can accommodate hardcopy documents of varying thickness. Once the overlapping images are recorded by the image acquisition system, the frame merger module assembles a composite image by identifying the region of overlap between the overlapping images. The composite image is subsequently transmitted for display on a standalone device or as part of a video conferencing system.

Dual video camera system for scanning hardcopy documents

A method carried out in an image processing system in which images of documents are captured by an image capture device, such as a video camera, comprising: (a) displaying successive images captured by the video camera, each image being defined by grayscale image data and containing text matter, (b) receiving a first user input (mouse button click) defining the start of a selection and a first position within the displayed image, (c) in response to the first user input, freezing the displayed image, (d) determining the skew angle of text matter with respect to the field of view of the video camera, (e) receiving at least one further user input (further button click; drag of cursor), including a final user input (mouse button release), defining the end of a selection, and for the or each further user input, (f) determining, using the skew angle determined in step (d), the position, shape and dimensions of a selection element in dependence upon at least the first position, and (g) displaying the selection element superimposed on the frozen displayed image. The selection element may be a rectangle, or a selection block highlighting one or more words of text.

Method and apparatus for selecting text and image data from video images

An image processing method and system for generating enhanced binary image data from greyscale input image data. The method includes the steps of (a) receiving first image data, the first image data being greyscale image data defining an input image, (b) performing a high frequency boost operation on the first image data to produce second image data, (c) performing a linear interpolation operation on the second image data to produce third image data, the third image data having a resolution higher than the resolution of the second image data, (d) performing a contrast enhancement operation on the third image data to produce fourth image data, and (e) thresholding the fourth image data to produce fifth image data, the fifth image data being binary image data defining an output image. The techniques find application, for example, in over-the-desk scanning of documents, and in video-conferencing.

Growing community of inventors

Guildford, United Kingdom

Michael James Taylor