Loading...
Please wait, while we are loading the content...
Similar Documents
Conclusions and Future Work 6.1 Conclusions
| Content Provider | Semantic Scholar |
|---|---|
| Abstract | Video services on multimedia networks are emerging, as technology progresses in related fields, such as video compression, video transmission, video storage, video compositing/manipulation, and multimedia authoring. This dissertation focuses on video signal compositing and manipulation. Our goals are to thoroughly investigate various issues related to video compositing and manipulation and to provide efficient solutions for low-cost high-performance video compositing systems. We recognized three different degrees of freedom in implementing video compositing: feature, location, and data format. We also observed strong interactions among these degrees of freedom. In many cases, we can compromise performance in one degree of freedom and gain great improvement in another. It is a non-trivial problem to find the optimal solution for video compositing for different specific applications. We need to explore the advantages of various approaches in each degree of freedom before optimizing the overall system performance. In Chapter 2, we explored the first degree of freedom— feature. We studied the basic operations involved in various video compositing/manipulation functions. The compositing functions can range from simple operations like opaque overlapping to complicated ones like pixel-wise manipulations of arbitrarily-shaped video objects. In order to provide a flexible and efficient platform for representing multi-point distributed video compositing, we also proposed a structured video model, in which video objects are defined to represent logical or physical components in a video scene. Compositing functions are used to process each individual video object or to combine multiple video |
| File Format | PDF HTM / HTML |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |