Digital Video Content Fingerprinting Based on Scale Invariant Interest Region Detection with an Array of Anisotropic Filters
First Claim
1. A method for content based video sequence identification comprising:
- determining active regions in video frames of a video sequence;
selecting video frames in response to temporal statistical characteristics of the determined active regions; and
using a two pass analysis to detect a set of initial interest points and interest regions in the selected video frames to reduce the effective area of images that are refined by complex filters that provide accurate region characterizations resistant to image distortion for identification of the video frames in the video sequence.
14 Assignments
0 Petitions
Accused Products
Abstract
Video sequence processing is described with various filtering rules applied to extract dominant features for content based video sequence identification. Active regions are determined in video frames of a video sequence. Video frames are selected in response to temporal statistical characteristics of the determined active regions. A two pass analysis is used to detect a set of initial interest points and interest regions in the selected video frames to reduce the effective area of images that are refined by complex filters that provide accurate region characterizations resistant to image distortion for identification of the video frames in the video sequence. Extracted features and descriptors are robust with respect to image scaling, aspect ratio change, rotation, camera viewpoint change, illumination and contrast change, video compression/decompression artifacts and noise. Compact, representative signatures are generated for video sequences to provide effective query video matching and retrieval in a large video database.
73 Citations
34 Claims
-
1. A method for content based video sequence identification comprising:
-
determining active regions in video frames of a video sequence; selecting video frames in response to temporal statistical characteristics of the determined active regions; and using a two pass analysis to detect a set of initial interest points and interest regions in the selected video frames to reduce the effective area of images that are refined by complex filters that provide accurate region characterizations resistant to image distortion for identification of the video frames in the video sequence. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18)
-
-
19. A method for interest point descriptor and signature generation, the method comprising:
-
generating refined interest point parameter vectors, wherein each vector contains a set (sx, sy, x, y, peak polarity); drawing a rectangular box encompassing Nsx×
Msy pixels and centered at an interest point (x, y) location associated with a set (sx, sy, x, y, peak polarity); anddetermining a multi-dimensional descriptor and a multi-dimensional signature based on pixels in the rectangular box. - View Dependent Claims (20, 21, 22, 23, 24, 25, 26, 27)
-
-
28. A method of applying an array of anisotropic filters for image scale-space analysis, the method comprising:
-
forming an array of elliptic-shaped, sampled anisotropic second order partial derivative Gaussian filters with (sx, sy) scale parameters; forming an array of rectangular-shaped, sampled anisotropic second order partial derivative Gaussian filters with finite spatial support with (sx, sy) scale parameters to approximate the elliptic-shaped sampled anisotropic second order partial derivative Gaussian filters; analyzing scale space for a selected frame by the array of rectangular sampled anisotropic second order partial derivative Gaussian filters of finite spatial support corresponding to different (sx, sy) scales; and convolving a region of pixels in a surrounding neighborhood of an initial interest point with the array of rectangular sampled anisotropic second order partial derivative Gaussian filters to produce a result that is resistant to image distortion. - View Dependent Claims (29, 30, 31, 32)
-
-
33. A computer readable medium storing a computer program which causes a computer system to perform a method of applying an array of anisotropic filters for image scale-space analysis, the method comprising:
-
forming an array of elliptic-shaped, sampled anisotropic second order partial derivative Gaussian filters with (sx, sy) scale parameters; forming an array of rectangular-shaped, sampled anisotropic second order partial derivative Gaussian filters with finite spatial support with (sx, sy) scale parameters to approximate the elliptic-shaped sampled anisotropic second order partial derivative Gaussian filters; analyzing scale space for a selected frame by the array of rectangular sampled anisotropic second order partial derivative Gaussian filters of finite spatial support corresponding to different (sx, sy) scales; and convolving a region of pixels in a surrounding neighborhood of an initial interest point with the array of rectangular sampled anisotropic second order partial derivative Gaussian filters to produce a result that is resistant to image distortion. - View Dependent Claims (34)
-
Specification