Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

DCT-based video downscaling transcoder using split and merge technique.

Kai-Tat Fung1, Wan-Chi Siu

  • 1Centre for Multimedia Signal Processing, Department of Electronic and Information Engineering, The Hong Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong. enktfung@eie.polyu.edu.hk

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|February 17, 2006
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

AnlightenDiff: Anchoring Diffusion Probabilistic Model on Low Light Image Enhancement.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2024
Same author

See360: Novel Panoramic View Interpolation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2022
Same author

Features Guided Face Super-Resolution via Hybrid Model of Deep Learning and Random Forests.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2021
Same author

Online-Learning-Based Bayesian Decision Rule for Fast Intra Mode and CU Partitioning Algorithm in HEVC Screen Content Coding.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2019
Same author

Clustering-Based Compression for Population DNA Sequences.

IEEE/ACM transactions on computational biology and bioinformatics·2017
Same author

Compression of Multiple DNA Sequences Using Intra-Sequence and Inter-Sequence Similarities.

IEEE/ACM transactions on computational biology and bioinformatics·2015
Same journal

Change-Prior-Guided Unsupervised Change Detection of Heterogeneous Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

AgonicDreamer: Enhancing Multi-View Consistency in Text-to-3D Generation via Rectified Score Distillation.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

BiCM-Prompt: Bidirectional Cross-Modal Prompt Tuning for Class-Incremental Learning on Multisource Remote Sensing Images.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

GoP-based Quality Enhancement on Video Compression.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Align then Tensorize: Multi-Level Consistent Anchor Graph Learning for Scalable Multi-View Clustering.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
Same journal

Beyond Fidelity: Diverse Image Synthesis via Retrieval-Augmented Diffusion.

IEEE transactions on image processing : a publication of the IEEE Signal Processing Society·2026
See all related articles

This study introduces a novel video downscaling transcoder that operates in the discrete cosine transform (DCT) domain, significantly reducing computational load. The new method achieves high-quality video output with substantial computational savings for real-time streaming.

Area of Science:

  • Video compression and processing
  • Digital signal processing
  • Computer vision

Background:

  • Conventional video downscaling requires computationally intensive decompression, pixel-domain downscaling, and recompression.
  • Downscaling in the discrete cosine transform (DCT) domain is challenging due to motion vector resampling mismatches and potential quality degradation.

Purpose of the Study:

  • To propose a new video transcoder architecture for efficient DCT-domain downscaling.
  • To address the mismatch between downsampled DCT coefficients and resampled motion vectors.
  • To achieve significant computational savings and high video quality in real-time video services.

Main Methods:

  • A novel DCT-domain video transcoder architecture utilizing a split and merge technique.
  • Dividing macroblocks into dominant and boundary regions for adaptive transcoding.

Related Experiment Videos

  • Transcoding the dominant region in the DCT domain and adaptively processing the boundary region.

    • Main Results:

    • Achieved significant computational savings compared to conventional methods.
    • Maintained high video quality with controlled re-encoding errors.
    • Demonstrated dynamic control over re-encoding errors in the boundary region.

    Conclusions:

    • The proposed DCT-domain video downscaling transcoder offers substantial computational efficiency.
    • The split and merge technique effectively handles motion vector resampling issues.
    • This architecture is suitable for video servers needing real-time, high-quality streaming to diverse clients.