公开(公告)号：US20230206632A1

公开(公告)日：2023-06-29

申请号：US17560386

申请日：2021-12-23

申请人： YAHOO AD TECH LLC

发明人： Deven Santosh SHAH ,  Avijit SHAH ,  Topojoy BISWAS ,  Biren BARODIA

IPC分类号： G06V20/40 ,  G06V20/70 ,  G06V10/764

CPC分类号： G06V20/44 ,  G06V20/70 ,  G06V10/764 ,  G06V20/41

摘要： The disclosed systems and methods provide a novel framework that enables cost-effective, accurate and scalable detection and recognition of key events in sporting or live events. The framework functions by creating a domain-specific video dataset with frame level annotations (i.e., deep domain datasets) and then training a lightweight camera view classifier to detect camera views for a given video. The disclosed framework uses pre-trained pose estimation and panoptic segmentation models along with geometric rules as labeling functions to define scene types and derive frame level classification training data. According to some embodiments, disclosed frameworks may be used to identify key persons or events, select a thumbnail corresponding to a key person or event, generate personalized highlights to enhance user experience and social media promotions for a team, sport or players, and predict and select the best camera view sequence for automatic highlights generation.

公开(公告)号：US20240296291A1

公开(公告)日：2024-09-05

申请号：US18662775

申请日：2024-05-13

申请人： YAHOO AD TECH LLC

发明人： Deven Santosh SHAH ,  Sukanya MOORTHY ,  Topojoy BISWAS

IPC分类号： G06F40/30 ,  G06F40/166 ,  G06F40/242 ,  G06F40/284 ,  G06F40/40 ,  G06N3/08

CPC分类号： G06F40/30 ,  G06F40/166 ,  G06F40/40 ,  G06N3/08 ,  G06F40/242 ,  G06F40/284

摘要： The example embodiments are directed toward improvements in document classification. In an embodiment, a method is disclosed comprising generating a set of sentences based on a document; predicting a set of labels for each sentence using a multi-label classifier, the multi-label classifier including a self-attended contextual word embedding backbone layer, a bank of trainable unigram convolutions, a bank of trainable bigram convolutions, and a fully connected layer the multi-label classifier trained using a weakly labeled data set; and labeling the document based on the set of labels. The various embodiments can target multiple use cases such as identifying related entities, trending related entities, creating ephemeral timeline of entities, and others using a single solution. Further, the various embodiments provide a weakly supervised framework to train a model when a labeled golden set does not contain a sufficient number of examples.

公开(公告)号：US20230161964A1

公开(公告)日：2023-05-25

申请号：US17534502

申请日：2021-11-24

申请人： YAHOO AD TECH LLC

发明人： Deven Santosh SHAH ,  Sukanya MOORTHY ,  Topojoy BISWAS

IPC分类号： G06F40/30 ,  G06F40/166 ,  G06F40/40 ,  G06N3/08

CPC分类号： G06F40/30 ,  G06F40/166 ,  G06F40/40 ,  G06N3/08 ,  G06F40/242

摘要： The example embodiments are directed toward improvements in document classification. In an embodiment, a method is disclosed comprising generating a set of sentences based on a document; predicting a set of labels for each sentence using a multi-label classifier, the multi-label classifier including a self-attended contextual word embedding backbone layer, a bank of trainable unigram convolutions, a bank of trainable bigram convolutions, and a fully connected layer the multi-label classifier trained using a weakly labeled data set; and labeling the document based on the set of labels. The various embodiments can target multiple use cases such as identifying related entities, trending related entities, creating ephemeral timeline of entities, and others using a single solution. Further, the various embodiments provide a weakly supervised framework to train a model when a labeled golden set does not contain a sufficient number of examples.