Abstract: The present disclosure relates to inferring device rendered object interaction behavior. An apparatus is configured to receive historical device rendered object interaction data associated with a client device. The historical device rendered object interaction data includes a plurality of records. Each record is associated with a device rendered object category. For each record, the device rendered object category is extracted. For each device rendered object category, a penalty value is generated. The penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device. A penalty decision, used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device, is generated.

Abstract: The present disclosure relates to inferring device rendered object interaction behavior. An apparatus is configured to receive historical device rendered object interaction data associated with a client device. The historical device rendered object interaction data includes a plurality of records. Each record is associated with a device rendered object category. For each record, the device rendered object category is extracted. For each device rendered object category, a penalty value is generated. The penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device. A penalty decision, used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device, is generated.

Abstract: The present disclosure relates to inferring device rendered object interaction behavior. An apparatus is configured to receive historical device rendered object interaction data associated with a client device. The historical device rendered object interaction data includes a plurality of records. Each record is associated with a device rendered object category. For each record, the device rendered object category is extracted. For each device rendered object category, a penalty value is generated. The penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device. A penalty decision, used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device, is generated.

Abstract: The present disclosure relates to inferring device rendered object interaction behavior. An apparatus is configured to receive historical device rendered object interaction data associated with a client device. The historical device rendered object interaction data includes a plurality of records. Each record is associated with an device rendered object category. For each record, the device rendered object category is extracted. For each device rendered object category, a penalty value is generated. The penalty value is a programmatically generated predicted network time delay during which device rendered objects associated with the device rendered object category are not to be transmitted to the client device. A penalty decision, used to determine whether to apply the penalty value to future transmissions of device rendered objects to the client device, is generated.

Abstract: Retrieval of images of objects from a large-scale database of object images, based on a query image. The database may, for example, contain images of objects such as faces, vehicles, people and luggage. Semantic attributes such as doors or windows in the case of vehicles are used as high level semantic cues to determine identities of objects in the images. Salient visual characteristics of the images are labeled with attribute information, and a transformation is learned so as to transform the labeled visual characteristics into a discrimination vector that discriminates between the labels. A similarity metric is learned using the discrimination vectors, such that different images depicting the same object are determined to be close while those having different objects are determined to be far apart. Candidates are retrieved based on a query image, and a re-ranking step may be applied to improve results. Validation experiments are described.

Abstract: Retrieval of images of objects from a large-scale database of object images, based on a query image. The database may, for example, contain images of objects such as faces, vehicles, people and luggage. Semantic attributes such as doors or windows in the case of vehicles are used as high level semantic cues to determine identities of objects in the images. Salient visual characteristics of the images are labeled with attribute information, and a transformation is learned so as to transform the labeled visual characteristics into a discrimination vector that discriminates between the labels. A similarity metric is learned using the discrimination vectors, such that different images depicting the same object are determined to be close while those having different objects are determined to be far apart. Candidates are retrieved based on a query image, and a re-ranking step may be applied to improve results. Validation experiments are described.