Systems and methods are described for training machine learning models to detect objects in image or video data. A system may select a first sample set of frames from one or more video files. Indications of a location of an object of interest in each of at least two sample frames may be received, then the system may determine the location of the object of interest across a number of intermediary frames using a tracker. Annotation data may be stored identifying the objects of interest in the sample frames, and the annotation data may be used in training a machine learning model to identify the object of interest in subsequently provided image or video data.

Systems and methods are provided for analyzing images or video using computer vision. Data comprising real time or near real time information or historical information is retrieved that is associated with a sporting event at a physical location. A time segment is identified of a display device at the physical location for acquisition. The display device is configurable to present visual sponsorship data during the time segment for an assigned sponsor. It is determined that one or more rules are satisfied by the data. An indication is transmitted that the first rule is satisfied to a computing device of a sponsor. A bid or valuation is generated based at least on the first rule being satisfied. A request to acquire the time segment is received from the computing device of the sponsor, and the display device at the physical location is caused to present visual sponsorship data for the sponsor during the time segment.

Systems and methods are provided for analyzing images or video using computer vision. Media items, including images or videos, are retrieved from one or more media channels, such as social media network services, streaming media networks, or broadcast networks. A media item is identified that depicts a object associated with a sponsor. A media cost equivalent is determined that is associated with the media item as distributed by the media channels. Using computer vision, a percentage is determined for the media cost equivalent. The percentage is determined based at least in part on at least one of size, clarity, duration or position of the sponsor object or another object in the media item. A sponsor valuation is calculated based at least in part on the media cost equivalent adjusted by the percentage.

Systems and methods are described for augmenting video data based on automated identification of one or more objects depicted in the video data. One or more classification models may identify an object of interest in video data. An aggregated duration count may be maintained that reflects a length of time that the object of interest has been depicted in the video data. This duration or additional metric data derived in part from the duration may be displayed in association with display of the video data and continuously updated during playback of the video data.

Provided are techniques for image text analysis for identifying hidden text. An Optical Character Reader (OCR) is utilized to extract a text string from an image. Context within the image is analyzed. It is determined that the extracted text string is a partial text string based on the context. For a first radius level of a plurality of radius levels, a segmented sub-image is identified around the partial text string within the first radius level, an image search on the segmented sub-image is performed to identify a candidate text string, and, in response to determining that the candidate text string is a complete text string, the complete text string is provided for performing an action.

Examples disclosed herein relate to image recognition instructions to receive a plurality of image frames, route each of the plurality of image frames to at least one of a plurality of element detectors, determine whether the respective one of the plurality of element detectors has recognized an embedded element and, in response to determining that the respective one of the plurality of element detectors has recognized the embedded element, cause a resource associated with the recognized embedded element to be retrieved.

An image processing apparatus includes an obtaining unit, a first classification unit, and a second classification unit. The obtaining unit obtains plural types of document images. The first classification unit classifies each of the plural types of document images obtained by the obtaining unit into any of classification items included in one category among plural categories, the plural categories each including plural classification items. The second classification unit classifies, in a case where there is a document image that is not classified into the one category by the first classification unit among the plural types of document images, the document image, which is not classified into the one category, into another category that includes at least one of classification items into which some of the document images are classified by the first classification unit.

Apparatus and methods are disclosed for simultaneously tracking multiple finger and palm contacts as hands approach, touch, and slide across a proximity-sensing, multi-touch surface. Identification and classification of intuitive hand configurations and motions enables unprecedented integration of typing, resting, pointing, scrolling, 3D manipulation, and handwriting into a versatile, ergonomic computer input device.

Disclosed are music symbol recognition apparatuses and methods that recognise music symbols from handwritten music notations. Various implementations may process handwritten music notations by segmenting the handwritten music notations into a plurality of elementary ink segments and then grouping the segments into graphical objects based on spatial relationships between the segments. One or more candidate music symbols may be determined for each graphical object, along with a symbol cost for each symbol, which represents a likelihood that the graphical object belongs to a predetermined class of symbols. The music symbol candidates may be parsed to form graphs based on grammar rules, and the graph most likely to represent the handwritten music notations may be selected for display or other use. The selection may be based on the symbol costs associated with each candidate and on spatial costs associated with the grammar rules that are applied to the candidates.

Apparatus and methods are disclosed for simultaneously tracking multiple finger and palm contacts as hands approach, touch, and slide across a proximity-sensing, multi-touch surface. Identification and classification of intuitive hand configurations and motions enables unprecedented integration of typing, resting, pointing, scrolling, 3D manipulation, and handwriting into a versatile, ergonomic computer input device.