摘要:
Methods are disclosed for finding images from a large corpus of images that at least partially match a query image. The present method makes use of feature detectors to bundle features into local groups or bundles. These bundled features are repeatable and much more discriminative than an individual SIFT feature. Equally importantly, the bundled features provide a flexible representation that allows simple and robust geometric constraints to be efficiently enforced when querying the index.
摘要:
This document relates to compositions and methods for modulating an immune response. For example, compositions of immunostimulatory CpG oligonucleotides derived from retroviral genomes are provided.
摘要:
Described is a technology in which face alignment data is obtained by processing an image using a component-based discriminative search algorithm. For each facial component, the search is guided by an associated directional classifier that determines how to move the facial component (if at all) to achieve better alignment relative to its corresponding facial component in the image. Also described is training of the classifiers.
摘要:
A swing-type submersible floating aerator, including a gas-ring compressor, a base, a pipe, a support tube, a swing device and a driving device, wherein the base is disposed between the gas-ring compressor and the support tube, the base is disposed at the center of the swing device, and the driving device drives the swing device to rotate. The swing-type submersible floating aerator is capable of rotating 360 degrees in a clockwise or a counterclockwise direction, and push-flow and aeration can be performed in any direction.
摘要:
Systems and methods provide picture collage systems and methods. In one implementation, a system determines a salient region in each of multiple images and develops a Bayesian model to maximize visibility of the salient regions in a collage that overlaps the images. The Bayesian model can also minimize blank spaces in the collage and normalize the percentage of each salient region that can be visibly displayed in the collage. Images are placed with diversified rotational orientation to provide a natural artistic collage appearance. A Markov Chain Monte Carlo technique is applied to the parameters of the Bayesian model to obtain image placement, orientation, and layering. The MCMC technique can combine optimization proposals that include local, global, and pairwise samplings from a distribution of state variables.
摘要:
A web-based computational system and method for high-throughput gene mining and target identification and validation based on conserved functional blocks automates the generation of conserved regions of high sequence similarity using algorithms and optimizes probe design for fragment amplification from homologous genomes.
摘要:
Cutout and/or keep away regions are etched in the winding structure near the gapped center leg of a magnetic core. This reduces eddy current losses caused by the leakage field and improves current uniformity and current sharing between windings thereby increasing converter efficiency. Windings closest to the air gap are suitably formed with both keep away regions and cutouts. Windings a little further away are formed with only cutouts, and the windings furthest away are unchanged. This approach keeps the net winding losses low in the presence of a fringing field. The precise configuration is determined by the core structure, air gap and winding arrangements.
摘要:
A vertical winding structure for planar integrated magnetics used in switched-mode power converters maintains close coupling between the different windings but reduces the eddy current losses, lowers the DC winding resistance and reduces the number of layers of the PCB. Vertical and horizontal windings can be used together without sacrificing these performance advantages and further minimizing the capacitive coupling between the outer-leg windings and the center-leg winding. This winding structure can be used in a wide range of magnetic structures including isolated and non-isolated CDRs, interleaved CDRs, and buck and boost converters.
摘要:
Digital video effects are described. In one aspect, a foreground object in a video stream is identified. The video stream comprises multiple image frames. The foreground object is modified by rendering a 3-dimensional (3-D) visual feature over the foreground object for presentation to a user in a modified video stream. Pose of the foreground object is tracked in 3-D space across respective ones of the image frames to identify when the foreground object changes position in respective ones of the image frames. Based on this pose tracking, aspect ratio of the 3-D visual feature is adaptively modified and rendered over the foreground object in corresponding image frames for presentation to the user in the modified video stream.
摘要:
A system that provides binds or associates a clickable hyperlink with an object that appears in a video stream. The hyperlink may be sent in a separate stream from the video stream, and user interfaces are provided to a user to activate the hyperlink. Activation of the hyperlink may cause a redirection to an associated website. Furthermore, feedback may be provided as to the user's activity regarding interest and activation regarding particular hyperlinks in the video.