Abstract: A system and method for facilitating high-density interactive voting (such as during a live event) using a computer network and tabulating and providing voting results in real-time. In general, the high-density interactive voting system of the present invention includes five major components that reside on the voting network: (1) a Live Event Wizard; (2) a Live Event Staging Component; (3) a Live Event Database; (4) a Live Event Object (LEO); and (5) a Live Event Display Engine. The Live Event Wizard is enables a user to quickly and easily create polling questions for distribution and presentation to a voter. The Live Event Staging Component ensures the validity of the polling questions and sends a copy of a live event definition and the polling questions to the Live Event Database. The Live Event Database tabulates current voting results at specified (or pre-determined) time intervals and performs statistical calculations of voting results.

Abstract: A system and process for broadcast and communication with bi-level or sketch video at extremely low bandwidths is described. Essentially, bi-level and sketch video presents the outlines of the objects in a scene being depicted. Bi-level and sketch video provides a clearer shape, smoother motion, shorter initial latency and cheaper computational cost than do conventional DCT-based video compression methods. This is accomplished by converting each color or gray-scale image frame to bi-level or sketch image frame using adaptive thresholding method, compressing bi-level or sketch image frames into bi-level or sketch video using adaptive context-based arithmetic coding method. Bi-level or sketch video is particularly suitable to such small devices as Pocket PCs and mobile phones that possess small display screen, low bandwidth connection, and light computational power.

Abstract: A “speech onset detector” provides a variable length frame buffer in combination with either variable transmission rate or temporal speech compression for buffered signal frames. The variable length buffer buffers frames that are not clearly identified as either speech or non-speech frames during an initial analysis. Buffering of signal frames continues until a current frame is identified as either speech or non-speech. If the current frame is identified as non-speech, buffered frames are encoded as non-speech frames. However, if the current frame is identified as a speech frame, buffered frames are searched for the actual onset point of the speech. Once that onset point is identified, the signal is either transmitted in a burst, or a time-scale modification of the buffered signal is applied for compressing buffered frames beginning with the frame in which onset point is detected. The compressed frames are then encoded as one or more speech frames.

Abstract: A local bi-gram model object recognition system and method for constructing a local bi-gram model and using the model to recognize objects in a query image. In a learning phase, the local bi-gram model is constructed that represents objects found in a set of training images. The local bi-gram model is a local spatial model that only models the relationship of neighboring features without any knowledge of their global context. Object recognition is performed by finding a set of matching primitives in the query image. A tree structure of matching primitives is generated and a search is performed to find a tree structure of matching primitives that obeys the local bi-gram model. The local bi-gram model can be found using unsupervised learning. The system and method also can be used to recognize objects unsupervised that are undergoing non-rigid transformations for both object instance recognition and category recognition.

Abstract: A multiple protection group (MPG) erasure-resilient coding method for constructing MPG codes for encoding and decoding data. The MPG codes constructed herein protect data chunks of data in multiple protection groups and subgroups. In general, the MPG erasure-resilient codes are constructed by locating data chunks into multiple protection groups and assigning at least one parity chunk to each protection group. Basic MPG codes are constructed from existing Maximum Distance Separable (MDS) codes by splitting at least some of the parity chunks into local parities for each of the multiple protection groups and projecting local parities onto each of the groups. Generalized MPG codes have a Maximally Recoverable property that can be used to determine whether an erasure pattern is recoverable or unrecoverable. Generalized MPG codes can recover any erasure pattern that is recoverable.

Abstract: The concurrent presentation technique provides information about content related to a source media currently being presented to a user in a fashion that allows the user to keep viewing the source media while either interactively or non-interactively perusing a list of related content. Thus, the user can see a list of related content without interrupting the presentation experience, and if desired, the user can choose to interact with the list to obtain further information about available related content.

Abstract: Images are automatically annotated using semantic distance learning. Training images are manually annotated and partitioned into semantic clusters. Semantic distance functions (SDFs) are learned for the clusters. The SDF for each cluster is used to compute semantic distance scores between a new image and each image in the cluster. The scores for each cluster are used to generate a ranking list which ranks each image in the cluster according to its semantic distance from the new image. An association probability is estimated for each cluster which specifies the probability of the new image being semantically associated with the cluster. Cluster-specific probabilistic annotations for the new image are generated from the manual annotations for the images in each cluster. The association probabilities and cluster-specific probabilistic annotations for all the clusters are used to generate final annotations for the new image.

Abstract: An “Oblique Image Stitcher” provides a technique for constructing a photorealistic oblique view from a set of input images representing a series of partially overlapping views of a scene. The Oblique Image Stitcher first projects each input image onto a geometric proxy of the scene and renders the images from a desired viewpoint. Once the images have been projected onto the geometric proxy, the rendered images are evaluated to identify optimum seams along which the various images are to be blended. Once the optimum seams are selected, the images are remapped relative to those seams by leaving the mapping unchanged at the seams and interpolating a smooth mapping between the seams. The remapped images are then composited to construct the final mosaiced oblique view of the scene. The result is a mosaic image constructed by warping the input images in a photorealistic manner which agrees at seams between images.

Abstract: A “Joint Bilateral Upsampler” uses a high-resolution input signal to guide the interpolation of a low-resolution solution set (derived from a downsampled version of the input signal) from low-to high-resolution. The resulting high-resolution solution set is then saved or applied to the original input signal to produce a high-resolution output signal. The high-resolution solution set is close to what would be produced directly from the input signal without downsampling. However, since the high-resolution solution set is constructed in part from a downsampled version of the input signal, it is computed using significantly less computational overhead and memory than a solution set computed directly from a high-resolution signal. Consequently, the Joint Bilateral Upsampler is advantageous for use in near real-time operations, in applications where user wait times are important, and in systems where computational costs and available memory are limited.

Abstract: A “Classifier Trainer” trains a combination classifier for detecting specific objects in signals (e.g., faces in images, words in speech, patterns in signals, etc.). In one embodiment “multiple instance pruning” (MIP) is introduced for training weak classifiers or “features” of the combination classifier. Specifically, a trained combination classifier and associated final threshold for setting false positive/negative operating points are combined with learned intermediate rejection thresholds to construct the combination classifier. Rejection thresholds are learned using a pruning process which ensures that objects detected by the original combination classifier are also detected by the combination classifier, thereby guaranteeing the same detection rate on the training set after pruning. The only parameter required throughout training is a target detection rate for the final cascade system.

Abstract: A photo illumination technology that facilitates user selectable illumination of an existing photo ranging from full artificial flash illumination to ambient light illumination and every gradation between. The technology in one embodiment employs a digital camera feature that captures two photos hundredths of a second apart where one photo is captured in concert with a flash and one is captured with no flash. Compensations are made to reconcile any movement of the subject matter that may have occurred between the two photos and the user is presented with the infinite gradations of illumination on the subject matter when selecting a photo for display or printing.

Abstract: A spatial-color Gaussian mixture model (SCGMM) image segmentation technique for segmenting images. The SCGMM image segmentation technique specifies foreground objects in the first frame of an image sequence, either manually or automatically. From the initial segmentation, the SCGMM segmentation system learns two spatial-color Gaussian mixture models (SCGMM) for the foreground and background objects. These models are built into a first-order Markov random field (MRF) energy function. The minimization of the energy function leads to a binary segmentation of the images in the image sequence, which can be solved efficiently using a conventional graph cut procedure.

Abstract: A “Vector Graphics Encoder” encodes vector graphics in a randomly accessible format. This encoding format enables particular portions of encoded images to be directly accessed, at any desired level of zoom, without processing or otherwise decoding the entire image. This random-access format is based on a coarse image grid of partially overlapping cells wherein each cell is defined by a “texel program.” Unlike fixed-complexity cells used by conventional vector images, each cell defined by a texel program is locally specialized without requiring global constraints on the complexity of each cell. The texel program for each cell is provided as a variable-length string of tokens representing a locally specialized description of one or more of layers of graphics primitives overlapping the cell. Images are then rendered by interpreting the texel programs defining one or more cells.

Abstract: An “Automatic Communicator” provides automated communications services between portable electronic devices (such as, for example, cell phones, PDA's, media players, portable computers, etc.) to facilitate information transfer and matching services while minimizing or eliminating the need for human involvement in responding to queries from one electronic device to another. Such communications are enabled by allowing portable electronic devices (“requesting devices”) to send a query to other portable electronic devices (“answering devices”), and have one or more of those answering devices automatically respond to the query without requiring users of the answering devices to become involved in providing that response. In effect, this allows the portable electronic devices of users to act as digital “butlers” or “personal assistants” that respond to queries on behalf of the users of those devices.

Abstract: A medical image acquisition error detection technique is presented which exploits the special characteristics of medical images to detect possible errors. In general, the presented technique categorizes medical images based on the type of images. For each category, a type of standard image is computed which represents the typical characteristics of images within a category. Then, exemplary images in a category are compared to the standard image for that category to establish an acceptable difference between actual images and the standard image for the category under consideration. When a medical image is to be examined for possible acquisition errors, it is categorized and a measure of difference between the image and the standard image for the category is computed. If the measure of difference falls outside the acceptable difference for the category, the image is deemed to contain an acquisition error and an alert is issued.

Abstract: An omni-directional camera (a 360 degree camera) is proposed with an integrated microphone array. The primary application for such a camera is videoconferencing and meeting recording, and the device is designed to be placed on a meeting room table. The microphone array is in a planar configuration, and the microphones are located as close to the desktop as possible to eliminate sound reflections from the table. The camera is connected to the microphone array base with a thin cylindrical rod, which is acoustically invisible to the microphone array for the frequency range [50-4000] Hz. This provides a direct path from the person talking to all of the microphones in the array, and can therefore be used for sound source localization (determining the location of the talker) and beam-forming (improving the sound quality of the talker by filtering only sound from a particular direction). The camera array is elevated from the table to provide a near frontal viewpoint of the meeting participants.

Abstract: The system provides two-sided local markets that enable local and/or intermittent resource allocation through a market model. The system provides a method for defining goods, services or assets are to be exchanged among a large but limited number of participants (e.g. between 10 and 100). In one embodiment, all participants may have both needs and endowments of the goods, services and assets to be traded and who may at times want to simultaneously buy and sell multiple units. Alternatively the market may have participants who are either sellers or buyers. The system provides a market mechanism where participants can submit orders (through limit orders and market orders) in user defined time frames.

Abstract: A system and process for dereverberation of multi-channel audio streams is presented which uses reverberation suppression techniques. In general, the present system and process builds a frequency dependent model of the reverberation decay and uses spectral subtraction-based reverberation reduction to achieve the aforementioned suppression. This dereverberation system and process can be used to improve automatic speech recognition (ASR) results with minimal CPU overhead.

Abstract: A “seamless multiplexer,” as described herein, provides a flexible and efficient system and method for multiplexing embedded coded bitstreams. Further, unlike the conventional JPEG 2000 standard, the seamless multiplexer is capable of decoding multiplexed encoded bitstreams without the need to use a demultiplexing aide as overhead in the encoded signal. In general, the seamless multiplexer provides a novel approach for multiplexing code block bitstreams by using a “decoder pointer,” for multiplexing multiple embedded bitstreams without the use of a demultiplexing aide. Consequently, the seamless multiplexer not only reduces any multiplexing overhead, but also allows much smaller granularity of access in reshaping the compressed input signal. In addition, in one embodiment, the seamless multiplexer also uses dependencies between coefficient blocks to further improve overall compression efficiency.

Abstract: The perceptual scalable audio coding/decoding technique lies in the use of a psychoacoustic mask to guide residue coding in enhancement layer coders. At the encoder, a psychoacoustic mask is calculated for the enhancement layer coders or is simply extracted from the coded base layer bitstream. One can also decode the coded base layer bitstream into the audio waveform, and calculate the psychoacoustic mask from the decoded base layer waveform. Furthermore, a predictive technology can be used to refine the psychoacoustic mask derived from the base layer bitstream to form a more accurate psychoacoustic mask of the enhancement layer. In addition, one can calculate the enhancement layer psychoacoustic mask from the original audio, and send the difference between the enhancement layer psychoacoustic mask and the base layer psychoacoustic mask as side information to the decoder. This psychoacoustic mask may then be used for the perceptual coding and decoding of the residue.