Video Search Engines

Real-Time Processing (Media Processing) (Video Search Engines)

For real-time processing applications, resource management is critical in order to achieve optimal performance. Distributed systems may be employed, but careful consideration of data bandwidth utilization is paramount. Event-based architectures can provide efficient system design and fit in well with publish/subscribe models [Wold02]. For user-contributed applications, we don’t necessarily have the issues associated with real-time […]

Systems Issues and Architectures (Media Processing) (Video Search Engines)

The media processing systems programmer can benefit from using one of the several frameworks that have been developed primarily for rendering and transcoding media of various formats on different platforms. Good modular systems design has led to implementation of dataflow architectures employing sources, sinks, and filters. These frameworks include the open source project gstreamer [Kata06], […]

Introduction to Video Processing (Video Search Engines)

Along with the expeditious improvements in the network, computation, storage technologies and consumer electronics, video content has become much more commonplace, and more accessible. Powered by the broadband wireless connection capabilities provided by Wi-Fi, 3G, WiMax, etc., and the video playback capabilities available on various personal digital assistants (PDA), the vision of searching and browsing […]

Shot Boundary Determination (Video Processing) (Video Search Engines) Part 1

Shot boundary determination (SBD) has been widely studied for the last decade. Some of the early work can be found in [Shahraray95, Wang00, Yeo95, Zhang93]. Researchers at AT&T started to tackle multimedia content processing and indexing in the early 1990s, and Shahraray reported a scene change detection algorithm in 1995 [Shahraray95]. With the limited computation […]

Shot Boundary Determination (Video Processing) (Video Search Engines) Part 2

Wipe Detector Wipe is the most ill defined transition. There are more than 20 different types of wipe that are commonly used in video editing and there is no single rule that applies to all of them. In this system, only one common type of wipe is considered, where the first scene gradually changes to […]

Representative Image Selection (Video Processing) (Video Search Engines)

Each video shot can be represented by a set of representative images. There are a few ways to select the representative images. The simplest one is to choose the first, the last, or the middle frame of the shot to be the representative image. But this choice may not be optimal. Although the criteria of […]

Face Detection (Video Processing) (Video Search Engines)

Face detection in still images has a wide range of applications, including image retrieval, multimodal human computer interfaces, multimedia content analysis, digital cameras, face recognition, face tracking, video surveillance, emotion detection, etc. In the last two decades, much research effort has been focused on this area, and a few successful face detection techniques have been […]

Face Recognition (Video Processing) (Video Search Engines)

The goal of face recognition is to recognize people by their facial features. Motivated by its wide area of applications, which include entertainment, human computer interfaces, law enforcement, biometric identification, and surveillance, etc., face recognition has attracted huge research attention and effort in the last few decades. Some of the advanced face recognition technologies, for […]

Video Optical Character Recognition (Video Processing) (Video Search Engines)

Optical character recognition (OCR) is a technique to translate images of handwritten or printed text into symbols. As a field of research, OCR benefited from the progress in pattern recognition, artificial intelligence, and computer vision. Over the last half contrary, OCR techniques have been constantly improved, and many commercial products have been successfully built. The […]

Concept Detection (Video Processing) (Video Search Engines)

Traditional image search leverages text associated with images, a low level content-based matching, or a combination of the two. A more intuitive content search needs to facilitate semantic concepts. The signal processing community has long studied low level features and derived high-level features (or semantic concepts) for large image databases. High-level concepts are generally learned […]

← Previous Entries

Next Entries →