Abstract: A document transcription application receives an image of a document that comprises structured data. The document transcription application performs optical character recognition upon the image of the document to produce a block of text. The document transcription application applies the block of text to a first machine learning model to determine a heat map for a class of data in the structured data in the image of the document. The document transcription application applies the image of the document and the heat map to a second machine learning model to identify a region of the image of the document representing the class of data. The document transcription application generates, using the identified region and the block of text, a structured data file.

Abstract: A collision warning system determines probabilities of potential collisions between a vehicle and other objects such as other vehicles. In an embodiment, sensors of a client device capture sensor data including motion data and image frames from a forward-facing view of the vehicle. An orientation of the client device relative to the vehicle may be determined using the motion data. The collision warning system determines cropped portions of the image frames and detects an object captured the image frames by processing the cropped portions. The collision warning system determines a probability of a potential collision between the vehicle and the object by tracking motion of the object. Responsive to determining that the probability is greater than a threshold value, the collision warning system may provide a notification of the potential collision to a driver of the vehicle.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: A collision warning system determines probabilities of potential collisions between a vehicle and other objects such as other vehicles. In an embodiment, sensors of a client device capture sensor data including motion data and image frames from a forward-facing view of the vehicle. An orientation of the client device relative to the vehicle may be determined using the motion data. The collision warning system determines cropped portions of the image frames and detects an object captured the image frames by processing the cropped portions. The collision warning system determines a probability of a potential collision between the vehicle and the object by tracking motion of the object. Responsive to determining that the probability is greater than a threshold value, the collision warning system may provide a notification of the potential collision to a driver of the vehicle.

Abstract: A method and device generates a trajectory. The method includes receiving a plurality of tracklets indicative of movement of a plurality of targets over a predetermined temporal interval. The method includes determining a plurality of context data for a pair of tracklets based upon at least one additional tracklet. The method includes computing a probability that the pair of tracklets relate to a first one of the targets. The method includes generating a trajectory for the first target based upon a concatenation of select ones of the tracklets. The concatenation maximizes the probability that the pair of tracklets correspond to the first target based upon the context data associated with the pair of the tracklets.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: A method and device generates a trajectory. The method includes receiving a plurality of tracklets indicative of movement of a plurality of targets over a predetermined temporal interval. The method includes determining a plurality of context data for a pair of tracklets based upon at least one additional tracklet. The method includes computing a probability that the pair of tracklets relate to a first one of the targets. The method includes generating a trajectory for the first target based upon a concatenation of select ones of the tracklets. The concatenation maximizes the probability that the pair of tracklets correspond to the first target based upon the context data associated with the pair of the tracklets.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Architecture that detects entrances on building facades. In a first stage, scene geometry is exploited and the multi-dimensional problem is reduced down to a one-dimensional (1D) problem. Entrance hypotheses are generated by considering pairs of locations along lines exhibiting strong gradients in the transverse direction. In a second stage, a rich set of discriminative image features for entrances is explored according to constructed designs, specifically focusing on properties such as symmetry and color consistency, for example. Classifiers (e.g., random forest) are utilized to perform automatic feature selection and entrance classification. In another stage, a joint model is formulated in three dimensions (3D) for entrances on a given facade, which enables the exploitation of physical constraints between different entrances on the same facade in a systematic manner to prune false positives, and thereby select an optimum set of entrances on a given facade.

Abstract: Multiple classifiers can be applied independently to evaluate images or video. Where there are heavily imbalanced class distributions, a local expert forest model for meta-level score fusion for event detection can be used. Performance variations of classifiers in different regions of a score space can be adapted. Multiple pairs of experts based on different partitions, or “trees,” can form a “forest,” balancing local adaptivity and over-fitting. Among ensemble learning methods, stacking with a meta-level classifier can be used to fuse an output of multiple base-level classifiers to generate a final score. A knowledge-transfer framework can reutilize the base-training data for learning the meta-level classifier. By recycling the knowledge obtained during a base-classifier-training stage, efficient use can be made of all available information, such as can be used to achieve better fusion and better overall performance.

Abstract: A method and device generates a trajectory. The method includes receiving a plurality of tracklets indicative of movement of a plurality of targets over a predetermined temporal interval. The method includes determining a plurality of context data for a pair of tracklets based upon at least one additional tracklet. The method includes computing a probability that the pair of tracklets relate to a first one of the targets. The method includes generating a trajectory for the first target based upon a concatenation of select ones of the tracklets. The concatenation maximizes the probability that the pair of tracklets correspond to the first target based upon the context data associated with the pair of the tracklets.

Abstract: Multiple classifiers can be applied independently to evaluate images or video. Where there are heavily imbalanced class distributions, a local expert forest model for meta-level score fusion for event detection can be used. Performance variations of classifiers in different regions of a score space can be adapted. Multiple pairs of experts based on different partitions, or “trees,” can form a “forest,” balancing local adaptivity and over-fitting. Among ensemble learning methods, stacking with a meta-level classifier can be used to fuse an output of multiple base-level classifiers to generate a final score. A knowledge-transfer framework can reutilize the base-training data for learning the meta-level classifier. By recycling the knowledge obtained during a base-classifier-training stage, efficient use can be made of all available information, such as can be used to achieve better fusion and better overall performance.