• Serif
  • Sans
 Star (-) Watch (-)

Research Notebook

Conditional Replenishment + Motion Compensation

Keywords

JPEG2000, Streaming, Conditional Replenishment, Motion Compensation

Conditional Replenishment of JPEG2000 tiles with Motion Compensation

Notes

  • Reduces the inter-frame redundancy present in common JPEG2000 sequences (MJP2 sequences).
  • Applies a real-time motion compensation technique to the MJP2 sequences before transmission.
  • They propose transmitting only the tiles that change in each JPEG2000 frame.
  • Review: the method proposed for real-time tile change detection.

References

  1. MIJ2K: Enhanced video transmission based on conditional replenishment of JPEG2000 tiles with motion compensation.
    Alvaro Luis Bustamante, José M. Molina López, Miguel A. Patricio.
    Univ. Carlos III de Madrid, Madrid, Spain.

Conditional Replenishment of JPEG2000 code-blocks with Motion Compensation

Notes

  • Too complex to be applied in real-time environments.
  • Clients and servers have to be purposely designed for use with these techniques.

References

  1. A novel paradigm for optimized scalable video transmission based on JPEG2000 with motion.
    A. Naman, D. Taubman.
    IEEE International Conference on Image Processing, 2007. ICIP 2007, vol. 5, 2007.

  1. Optimized scalable video transmission based on conditional replenishment of JPEG2000 code-blocks with motion compensation. A. Naman, D. Taubman.
    Proceedings of the International Workshop on Mobile Video, ACM, New York, NY, USA (2007), pp. 43–48

  2. A flexible video transmission system based on JPEG 2000 conditional replenishment with multiple references.
    F. Devaux, J. Meessen, C. Parisot, J. Delaigle, B. Macq, C. De Vleeschouwer.
    IEEE International Conference on Acoustics, Speech Signal Proces. (ICASSP 07).

Notes

  • They propose a rate-distortion optimal strategy to select the most profitable packets to transmit.
  • They provide the client with two references, the previous reconstructed frame and an estimation of the current scene background, which improves the transmission system performances.

References

Thesis

Notes

  • JSIV relies on three main concepts: storing the video sequence as independent JPEG2000 frames to provide for quality and spatial resolution scalability, as well as temporal and spatial accessibility; prediction and conditional replenishment of precincts to exploit inter-frame redundancy; and loosely-coupled server and client policies.

  • The server optimally selects the number of quality layers for each precinct it transmits and decides on any side-information that needs to be transmitted while the client attempts to make most of the received (distorted) frames.

  • The client decides which precincts are predicted and which are decoded from received data (or possibly filled with zeros in the absence of received data). Thus, in JSIV, a predicted frame typically has some of its precincts predicted from nearby frames while others are decoded from received intra-coded precincts; JSIV never uses frame differences or prediction residues.

  • The philosophy behind these policies is that neither the server nor the client drives the video streaming interaction, but rather the server dynamically selects and sends the pieces that, it thinks, best serve the client needs and, in turn, the client makes most of the pieces of information it has. The JSIV paradigm postulates that if both the client and the server policies are intelligent enough and make reasonable decisions, then the decisions made by the server are likely to have the expected impact on the client’s decisions.

  • Experimental results show that JSIV’s performance is slightly inferior to that of existing predictive coding standards in conventional streaming applications; however, JSIV produces significant improvements when its scalability and accessibility features, such as the region of interest, are employed.

  • The preprocessing stage is responsible for compressing each frame independently of the other frames into the JPEG2000 format and preparing side- information for these frames. The side information can include distortion-length slope tables, motion information, motion distortions, and any other side information that might be required during media serving. Side-information can either be generated off- line for pre-recorded media or in real-time for live media. Many of these operations are independent of each other and can be easily delegated to one or more machines in a content delivery network.

Patents