Зарегистрироваться
Восстановить пароль
FAQ по входу

Schonfeld D. et al. (Eds.) Video Search and Mining

  • Файл формата pdf
  • размером 13,80 МБ
  • Добавлен пользователем
  • Отредактирован
Schonfeld D. et al. (Eds.) Video Search and Mining
Springer, 2009. — 272 p. — (Studies in Computational Intelligence, Volume 287).
As cameras become more pervasive in our daily life, vast amounts of video data are generated. The popularity of YouTube and similar websites such as Tudou and Youku provides strong evidence for the increasing role of video in society. One of the main challenges confronting us in the era of information technology is to effectively rely on the huge and rapidly growing video data accumulating in large multimedia archives. Innovative video processing and analysis techniques will play an increasingly important role in resolving the difficult task of video search and retrieval. A wide range of video-based applications have benefited from advances in video search and mining including multimedia information management, human-computer interaction, security and surveillance, copyright protection, and personal entertainment, to name a few.
This book provides an overview of emerging new approaches to video search and mining based on promising methods being developed in the computer vision and image analysis community. Video search and mining is a rapidly evolving discipline whose aim is to capture interesting patterns in video data. It has become one of the core areas in the data mining research community. In comparison to other types of data mining (e.g. text), video mining is still in its infancy. Many challenging research problems are facing video mining researchers. For example, how to extract knowledge from spatio-temporal data? How to infer high-level semantic concepts from low-level features in videos? How to exploit unlabeled and untagged video data? The use of classical data mining techniques for video data is impractical due to the massive volume of high-dimensional video data. To address these difficult challenges, it is necessary to develop search and mining techniques and methods that are suitable for video data.
The objective of this book is to present the latest advances in video search and mining covering both theoretical approaches and practical applications. The book provides researchers and practitioners a comprehensive understanding of the startof- the-art in video search and mining techniques and a resource for potential applications and successful practice. This book can also serve as an important reference tool and handbook for researchers and practitioners in video search and mining.
The target audience of this book is mainly engineers and students working on video analysis in various disciplines, e.g. computer vision, pattern recognition, information technology, image processing, and artificial intelligence. The book is intended to be accessible to a broader audience including researchers and practicing professionals working in video applications such as video surveillance, video retrieval, etc.
The origin of this book stems from the immense success of the First International Workshop on Video Mining (VM’08), held in conjunction with the IEEE International Conference on Data Mining 2008. This workshop gathered experts from different fields working on video search and mining.
Part I: Motion Trajectory Analysis
Object Trajectory Analysis in Video Indexing and Retrieval Applications
Trajectory Clustering for Scene Context Learning and Outlier Detection
Motion Trajectory-Based Video Retrieval, Classification, and Summarization
Part II: High-Dimensional Video Representation
Three Dimensional Information Extraction and Applications to Video Analysis
Statistical Analysis on Manifolds and Its Applications to Video Analysis
Part III: Semantic Video Analysis
Semantic Video Content Analysis
Video Genre Inference Based on Camera Capturing
Visual Concept Learning from Weakly Labeled Web Videos
Part IV: Personalized Video
Face Recognition and Retrieval in Video
A Human-Centered Computing Framework to Enable Personalized News Video Recommendation
Part V: Video Mining
A Holistic, In-Compression Approach to Mining Independent Motion Segments for Massive Surveillance
Video Collections
Video Repeat Recognition and Mining by Visual Features
Mining TV Broadcasts 24/7 for Recurring Video YouTube Scale, Large Vocabulary Video Annotation
  • Чтобы скачать этот файл зарегистрируйтесь и/или войдите на сайт используя форму сверху.
  • Регистрация