Information Technology Reference
In-Depth Information
set
α
2
= 10. The results displayed in Fig. 16 indeed show that the two clips
within a sequence are discriminated independently of which degradation is applied
to the reference GOP.
α
1
= 1
,
4
Conclusion and Perspectives
This chapter has presented an overview of two frameworks for scalable video in-
dexing based on hierarchical decomposition. The first deals directly with the video
content in the compressed domain, i.e. the domain defined by the given compres-
sion standard, while the second relies on a hierarchical decomposition of the de-
coded data. Both approaches have their own advantages and drawbacks. Techniques
working in the compressed domain provide potentially faster processing since basic
features such as the motion vectors are readily available; however, the features are
computed so as to achieve the best quality-versus-compression ratio trade-off and
might not ideally serve the purpose of indexing. Techniques processing decoded
data, on the other hand, give full freedom to design ad-hoc multiscale descriptors;
however they are significantly computationally slower due to the full decoding fol-
lowed by the chosen hierarchical decomposition. Both approaches open an exciting
perspective for efficient processing of HD content such as indexing but also fast
browsing, summarizing and retrieval of HD video content by exploiting the various
levels of granularity that can be accessed in new information systems.
References
1. JPEG 2000 image coding system: Motion JPEG 2000, ISO/IEC. 15444-3:2007. Infor-
mation technology edn.
2. Morand, C., Benois-Pineau, J., Domenger, J.P., Mansencal, B.: Object-based indexing of
compressed video content: From sd to hd video. In: International Conference on Image
Analysis and Processing Workshops, Modena, Italy (2007)
3. Adami, N., Boschetti, A., Leonardi, R., Migliorati, P.: Embedded indexing in scalable
video coding. In: 7th International Workshop on Content-Based Multimedia Indexing,
Chania, Crete (2009)
4. Morand, C., Benois-Pineau, J., Domenger, J.P.: HD motion estimation in a wavelet
pyramid in JPEG 2000 context. In: 5th International Conference on Image Processing,
Chicago, IL, USA (2008)
5. JPEG 2000 image coding system: Core coding system, ISO/IEC 15444-1:2004. Infor-
mation technology edn.
2009)
7. Pearson, G., Gill, M.: An evaluation of Motion JPEG 2000 for video archiving. In:
Archiving, Washington, D.C., USA (2005)
8. Coding of audio-visual objects - Part 2: Visual (MPEG4), ISO/IEC 14496-2:2004. In-
formation technology edn.
9. Wang, Y., Hannuksela, M., Gabbouj, M.: Error-robust inter/intra mode selection using
isolated regions. In: Packet Video (PV), Nantes, France (2003)
