INTERFRAME CODING OF CANONICAL PATCHES FOR LOW BIT-RATE MOBILE AUGMENTED REALITY
| Autor: | Sam S. Tsai, David Chen, Vijay Chandrasekhar, Mina Makar, Bernd Girod |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Linguistics and Language
Computer Networks and Communications Computer science business.industry Detector ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Inter frame Computer Science Applications Image (mathematics) Artificial Intelligence Salient Feature (computer vision) Wireless Computer vision Augmented reality Artificial intelligence business Image retrieval Software Information Systems |
| Zdroj: | International Journal of Semantic Computing. :5-24 |
| ISSN: | 1793-7108 1793-351X |
| DOI: | 10.1142/s1793351x13400011 |
| Popis: | Local features are widely used for content-based image retrieval and augmented reality applications. Typically, feature descriptors are calculated from the gradients of a canonical patch around a repeatable keypoint in the image. In this paper, we propose a temporally coherent keypoint detector and design efficient interframe predictive coding techniques for canonical patches and keypoint locations. In the proposed system, we strive to transmit each patch with as few bits as possible by simply modifying a previously transmitted patch. This enables server-based mobile augmented reality where a continuous stream of salient information, sufficient for image-based retrieval and localization, can be sent over a wireless link at a low bit-rate. Experimental results show that our technique achieves a similar image matching performance at 1/15 of the bit-rate when compared to detecting keypoints independently frame-by-frame and allows performing streaming mobile augmented reality at low bit-rates of about 20–50 kbps, practical for today's wireless links. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|