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

Generalized face super-resolution.

Kui Jia1, Shaogang Gong

  • 1Shenzhen Institute of Advanced Integration Technology, Chinese Academy of Sciences/Chinese Academy of Hong Kong, Shenzhen, China. chrisjia@dcs.qmul.ac.uk

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society
|May 17, 2008
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

YOTO++: Learning Long-Horizon Closed-Loop Bimanual Manipulation from One-Shot Human Video Demonstrations.

IEEE transactions on pattern analysis and machine intelligence·2026
Same author

Multimodal prehabilitation enhances perioperative outcomes in gastric cancer patients: a single-center randomized controlled trial.

Frontiers in nutrition·2026
Same author

Surrogate Decision-Making by Family Caregivers for Hyperthermic Intraperitoneal Chemotherapy in Gastric Cancer: Qualitative Study in a High-Volume Chinese Center.

JMIR cancer·2026
Same author

Rehmannins A-D, Anti-inflammatory Carotenoid Pigments from the Fresh Root of <i>Rehmannia glutinosa</i>.

Organic letters·2026
Same author

Association of the C-reactive protein-triglyceride-glucose index with liver disease risk: findings from a nationwide Chinese cohort.

BMC gastroenterology·2026
Same author

Semi-Supervised Unconstrained Head Pose Estimation in the Wild.

IEEE transactions on pattern analysis and machine intelligence·2025
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 hierarchical tensor approach for generalized face super-resolution, enabling high-resolution image generation across various expressions and poses. The method outperforms existing techniques in both single and multimodal face hallucination tasks.

Area of Science:

  • Computer Vision
  • Machine Learning
  • Image Processing

Background:

  • Current face super-resolution methods are limited to single facial modalities (fixed expression, pose, illumination).
  • A need exists for generalized approaches that can handle variations in facial appearance.

Purpose of the Study:

  • To develop a generalized face super-resolution technique capable of hallucinating high-resolution images across multiple facial modalities (expression, pose).
  • To improve upon existing single-modal face hallucination methods.

Main Methods:

  • Utilized a hierarchical tensor (multilinear) space representation.
  • Formulated a unified tensor comprising a global image-based tensor for cross-modal mapping and a local patch-based multiresolution tensor for high-resolution details.

Related Experiment Videos

  • Developed an automatic face alignment algorithm for pixel-wise alignment of unregistered low-resolution faces.

    • Main Results:

    • Demonstrated superior performance compared to existing benchmark face super-resolution techniques on single-modal hallucination.
    • Showcased the approach's novelty in handling multimodal face hallucination.
    • Confirmed robustness in automatic alignment under practical imaging conditions.

    Conclusions:

    • The proposed hierarchical tensor approach offers a generalized solution for face super-resolution across multiple modalities.
    • The method achieves state-of-the-art performance and robust alignment for realistic face hallucination.