Automatic Naming of Speakers in Video via Name-Face Mapping

来源 :第十五届全国计算语言学学术会议(CCL2016)暨第四届基于自然标注大数据的自然语言处理国际学术研讨会(NLP-NABD | 被引量 : 0次 | 上传用户：yyy021

【摘要】

：

【作者】

：

Zhixin Liu Cheng Jin Yuejie Zhang Tao Zhang

【机构】

：

School of Computer Science Shanghai Key Laboratory of Intelligent Information Processing,Fudan Unive

【出处】

：

第十五届全国计算语言学学术会议(CCL2016)暨第四届基于自然标注大数据的自然语言处理国际学术研讨会(NLP-NABD

【发表日期】

：

2016年6期

【关键词】

：

Automatic Naming Speakers in Video Face and Clustering Extraction Name Extractio

下载到本地 , 更方便阅读

下载此文赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

　　The problem of automatically labelling the appearances of characters in video with their names is challenging due to the huge variation in the appearance of each character and the weakness and ambiguity of available annotations.We can achieve high precision by combining multiple sources of information,both visual and textual.The principal novelties that we introduce in this paper are:(i)extracting face features in video by neural network;(ii)strengthening the mapping between names and faces by analyzing the co-occurrence of names and faces;(iii)automatically and efficiently labelling appearances of main characters with their names.

其他文献

Semi-supervised Learning for Mongolian Morphological Segmentation

Unlike previous Mongolian morphological segmentation methods based on large labeled training data or complicated rules concluded by linguists,we explore a novel semi-supervised method for a practical

会议

Semi-supervised learningMorphological segmentationStatistical machine translat

Recognizing Biomedical Named Entities Based on the Sentence Vector/Twin Word Embeddings Conditioned

As a fundamental step in biomedical information extraction tasks,biomedical named entity recognition remains challenging.In recent years,the neural network has been applied on the entity recognition t

会议

LSTMtwin word embeddingssentence vectorViterbi algorithm

Combining Event-level and Cross-event Semantic Information for Event-Oriented Relation Classificatio

Previous researches on event relation classification primarily rely on lexical and syntactic features.In this paper,we use a Shallow Convolutional Neural Network(SCNN)to extract event-level and cross-

会议

Event Relation ClassificationSemantic InformationFrame EmbeddingSCNN

The Constitution of a Fine-Grained Opinion Annotated Corpus on Weibo

Sentiment analysis on social media represented by Weibo is one of the hotspot research problems in NLP.A comprehensive and systematic fine-grained annotated corpus plays a significance role.In this pa

会议

weibo corpusfine-grained opinion annotationimplicit opinion annotationevaluat

The Construction of a Customized Medical Corpus for Assisting Chinese Clinicians in English Research

A great number of clinicians in mainland China are under increasing pressure to publish their research results on international journals,and they urgently need support for writing research articles in

会议

Customized medical corpusChinese cliniciansresearch article writingCCUT

A New Focus Strategy for Efficient Dialog Management

The dialog manager is the most important component for a dialog system,in which the dialog state tracking is crucial to a real-world system.We claim that the intractability of dialog states comes from

会议

News Abridgement Algorithm Based on Word Alignment and Syntactic Parsing

The rapid development of new media results in a lot of redundant information,increasing the difficulty of quickly obtaining useful information and browsing simplified messages on portable devices.Thus

会议

News AbridgementKeyword FeaturesSentence AbridgementHeuristic RulesSentence

A Novel Approach for Discovering Local Community Structure in Networks

The algorithms for discovering global community structure require the knowledge about entire network structures,which are still difficult and unrealistic to obtain from nowadays extremely large networ

会议

Local Community DiscoveringCommunity StructureConnection DegreeNetwork Graph

Identifying Suspected Cybermob on Tieba

This paper describes an approach to identify suspected cybermob on social media.Many researches involve making predictions of group emotion on Internet(such as quantifying sentiment polarity),but this

会议

Netizen IdentificationSuspected CybermobMachine LearningSupport Vector Machin

Towards Scalable Emotion Classification in Microblog Based on Noisy Training Data

The availability of labeled corpus is of great importance for emotion classification tasks.Because manual labeling is too time-consuming,hashtags have been used as naturally annotated labels to obtain

会议

emotion classificationdata cleaninghashtagk-NN

Automatic Naming of Speakers in Video via Name-Face Mapping

其他学术论文