Dolby Laboratories Licensing Corporation

Wenyu Jiang

Claus Bauer

Regunathan Radhakrishnan

Tianshi Gao

Glenn N Dickins

Jeffrey Charles Riedmiller

David N Lathrop

Junfeng He

Regunthan Radhakrishnan

Attributes are identified in media content. A classification value of the media content is computed based on the identified attributes. Thereafter, a fingerprint derived from the media content is stored or searched for based on the classification value of the media content.

Storing and searching fingerprints derived from media content based on a classification of the media content

Techniques for re-associating dynamic metadata with media data are provided. A media processing system creates, with a first media processing stage, binding information comprising dynamic metadata and a time relationship between the dynamic metadata and media data. The binding information may be derived from the media data. While the first media processing stage delivers the media data to a second media processing stage in a first data path, the first media processing stage passes the binding information to the second media processing stage in a second data path. The media processing system re-associates, with the second media processing stage, the dynamic metadata and the media data using the binding information.

Alignment and re-association of metadata for media streams within a computing device

Fingerprint-based content identification is described. Bin-counting detects linear trends based on data points that are generated from a set of query fingerprints, which are extracted from a query content item and a matching set of corresponding reference content items that are extracted from a reference content item. RANSAC may detect trends based on confidence scores that are computed for multiple candidate content items. The trends detected for the candidate content items are compared against a threshold in order of decreasing confidence score until a trend that satisfies the threshold is found. A ranking-based mechanism for fingerprint-based content identification may use a bin-counting or RANSAC-based trend detection mechanism to identify duplicate or multiple reference content items that all match the same query content item.

Trend analysis in content identification based on fingerprinting

Derivation of a fingerprint includes generating feature matrices based on one or more training images, generating projection matrices based on the feature matrices in a training process, and deriving a fingerprint for one or more images by, at least in part, projecting a feature matrix based on the one or more images onto the projection matrices generated in the training process.

Scalable media fingerprint extraction

The present invention may be used to estimate operational characteristics of devices that transmit and receive streams of information in a communication system. In one application, the level of occupancy of a FIFO buffer in a processing device such as a router or wireless access point is estimated by monitoring packets transmitted by the processing device. Estimates of the operational characteristics can be used to control communications in the system so that the overall performance is improved. Techniques that can be used to mitigate effects of low signal-to-noise ratio conditions are also disclosed.

Estimating wireless processing device queue length and estimating signal reception quality in a wireless network

Content identification and quality monitoring are provided. The method involves obtaining a first fingerprint derived from a first media content, processing the first media content to generate a second media content, obtaining a second fingerprint derived from the second media content, and comparing the first fingerprint and the second fingerprint to determine one or more of: a similarity between the first fingerprint and the second fingerprint that indicates that the second media content is generated from the first media content or a difference between the first fingerprint and the second fingerprint to identify a quality degradation between the first media content and the second media content.

Content identification and quality monitoring

A value is computed for a feature in an instance of query content and compared to a threshold value. Based on the comparison, first and second bits in a hash value, which is derived from the query content feature, are determined. Conditional probability values are computed for the likelihood that quantized values of the first and the second bits equal corresponding quantized bit values of a target or reference feature value. The conditional probabilities are compared and a relative strength determined for the first and second bits, which directly corresponds to the conditional probability. The bit with the lowest bit strength is selected as the weakbit. The value of the weakbit is toggled to generate a variation of the query hash value. The query may be extended using the query hash value variation.

Non-parametric measurement of media fingerprint weak bits

Search tree structures with nodes that represent signatures derived from segments of video and audio content are used by systems to identify content and re-establish the correct alignment between video and audio content that have become disassociated with one another. The amount of storage needed to record data representing the tree structure can be reduced by replacing stored signature sets with signature pointers. The efficiency of searches in the tree structure can be improved by constructing and using partial tree structures.

Incremental construction of search tree with signature pointers for identification of multimedia content

A receiver monitoring the transmissions of a packet processing device such as a wireless access point analyzes the monitored transmissions to estimate the probability of packet loss due to buffer overflow in the processing device. The loss probability can be estimated for individual packets to follow rapid changes in communication conditions. An indication of the estimated loss probability can be used to adapt the forward error correction (FEC) parameters that are used to generate the packets, thereby improving the reliability of data delivery to a receiver while optimizing the use of communication channel bandwidth.

Transient analysis of packet queuing loss in a broadcast network

A method for use in identifying a segment of audio and/or video information comprises obtaining a query fingerprint at each of a plurality of spaced-apart time locations in said segment, searching fingerprints in a database for a potential match for each such query fingerprint, obtaining a confidence level of a potential match to a found fingerprint in the database for each such query fingerprint, and combining the results of searching for potential matches, wherein each potential match result is weighted by respective confidence level. A confidence level may be a function of at least one or both of (1) a measure of difference between a query fingerprint and a found fingerprint and (2) the relative timing relationship between the time location of a query fingerprint and the time location of a found fingerprint.

Audio/Video fingerprint search accuracy using multiple search combining

Bit errors in packets of data that are communicated in a network such as a wireless network can be corrected by processes that do not require any overhead in the data such as conventional error-detection codes or redundant information such as conventional error-correction codes. A validation-set process compares corrupted data against values in a set of known valid values and selects one of the known valid values to replace the corrupted data. A consistency-check process uses data correlation characteristics of two or more parameters to determine if values obtained from a packet are consistent with one another. If not, values are changed to make them consistent.

Growing community of inventors

Singapore, Singapore

Wenyu Jiang