Nvidia Corporation

Ati International Srl

Michael L Lightstone

Jonathan B White

Stefan Eckart

Richard W Webb

Xiaohua Yang

Fabio Ingrao

Haitao Guo

Hassane S Azar

Field-based detection of 3:2 pulldown in a sequence of digital video fields using a programmable graphics processor is described. The detection is performed using a threshold value to determine equivalence between a pair of fields of digital video data. Furthermore, additional threshold values may be used to control switching into a mode where duplicated fields of digital video data are identified and not displayed and out of the mode where duplicated fields of digital video data are identified and not displayed. Look ahead can be used to detect when to switch into or out of the mode where duplicated fields of digital video data are identified and not displayed, reducing the occurrence of visual artifacts.

3:2 pulldown detection

3:2 Pulldown detection

A software program includes at least two performance levels. Each performance level has an associated processor utilization. Each performance level corresponds to optimization criteria, such as a quality of data processing performed by the software program. The performance level is selected to maintain processor utilization by the software program within constraints, such as a desired range of processor utilization and a minimum idle thread allocation.

Managing a video encoder to facilitate loading and executing another program

A video encoder includes a programmable rate controller. In one embodiment, the programmable rate controller includes a variable bit rate controller, a constant bit rate controller, and an arbitration logic for selecting one of the two outputs. An embodiment of a variable bit rate controller tracks long-term changes to average bit rate. An embodiment of a constant bit rate controller classifies macroblock types, determines a statistical indicator of complexity for each macroblock type, and generates a target bit rate based on estimated complexity.

Apparatus, system, and method for video encoder rate control

The invention provide methods and code for better detecting 3:2 pulldown or other video formats. In one respect, embodiments of the invention improve the way in which fields of video data are compared. In another respect, embodiments of the invention provide pattern matching techniques and code for processing the field difference data.

Method for video format detection

A method and apparatus for storing multimedia data for use in a digital VCR includes processing that begins by maintaining a first link list of a plurality of memory sections, where the first link list links the plurality of memory sections as a circular buffer. The processing then continues by receiving a stream of multimedia data. The processing then continues by storing the stream of multimedia data in at least some of the memory sections of the plurality of memory sections. The processing then continues by receiving a request for independent storage of a selected portion of the stream of multimedia data, e.g., the user desires to have a permanent copy of a particular program. The processing further continues by generating a second link list for a set of memory sections of the at least some of the memory sections. The set of memory sections stores the selected portion of the stream of multimedia data. The processing continues by updating the linking of the plurality of memory sections within the first link list without the set of memory section.

Method and apparatus for storing multi-media data

A method and apparatus for indexing and locating key frames in streaming frame data and variable-frame-length data is described. Fast and efficient location of desired key frames in both directions (e.g., forward/backward, future/past) is provided. An estimate of the distance to the desired key frame is made and a seek performed according to that distance. At the location specified by the seek, key frame seek assist data are obtained. The desired key frame is located or a new seek is performed, depending on the key frame seek assist data obtained. By placing the key frame seek assist data at readily identifiable locations, efficient location of key frames is provided. The key frame seek assist data may be tailored to optimize efficiency for seeking in a particular direction, for example, backwards. The seek process is correspondingly configured to favor seeking in the more efficient direction.

Method and apparatus for indexing and locating key frames in streaming and variable-frame-length data

The present invention provides a method for improving playback consistency of an encoded video stream. The method of the present invention first determines whether a series of values of the progressive_frame flag of consecutive frames of the encoded video stream are arranged in a predetermined pattern. If the values are arranged in the predetermined pattern, an image of at least one of the frames is displayed progressively regardless of the value of the progressive_frame flag of that frame. As a result, the video stream is played back as if the movie had been consistently encoded in the first place.

Growing community of inventors

Fremont, CA, United States of America

Michael L Lightstone