Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Search research articles

Related Experiment Video

Updated: May 26, 2026

Quantification of Orofacial Phenotypes in Xenopus

Quantification of Orofacial Phenotypes in Xenopus

Published on: November 6, 2014

Multidimensional scaling for matching low-resolution face images.

Soma Biswas¹, Kevin W Bowyer, Patrick J Flynn

¹Department of Computer Science and Engineering, University of Notre Dame, 384 Fitzpatrick Hall of Engineering, Notre Dame, IN 46556, USA. sbiswas@nd.edu

IEEE Transactions on Pattern Analysis and Machine Intelligence

|December 28, 2011

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

Intergroup randomized phase III study of adjuvant FOLFIRI, FOLFOX, or 5-FU/leucovorin for stage II/III rectal cancer: ECOG-ACRIN E3201.

The oncologist·2025

Same author

Lessons from Collecting a Million Biometric Samples.

Image and vision computing·2024

Same author

How context shapes collective turnover over time: The relative impact of internal versus external factors.

The Journal of applied psychology·2024

Same author

A Cohort Study to Evaluate Genetic Predictors of Aromatase Inhibitor Musculoskeletal Symptoms: Results from ECOG-ACRIN E1Z11.

Clinical cancer research : an official journal of the American Association for Cancer Research·2024

Same author

Long-Term Follow-Up of the Anthracyclines in Early Breast Cancer Trials (USOR 06-090, NSABP B-46-I/USOR 07132, and NSABP B-49 [NRG Oncology]).

Journal of clinical oncology : official journal of the American Society of Clinical Oncology·2024

Same author

Endophytic bacteria Klebsiella spp. and Bacillus spp. from Alternanthera philoxeroides in Madiwala Lake exhibit additive plant growth-promoting and biocontrol activities.

Journal, genetic engineering & biotechnology·2023

This study introduces a new method using Multidimensional Scaling (MDS) to improve face recognition with low-resolution (LR) images. The approach effectively matches low-resolution faces to high-resolution ones, enhancing accuracy in surveillance and distant imaging scenarios.

Area of Science:

Computer Vision
Biometrics
Machine Learning

Background:

Face recognition systems struggle with low-resolution (LR) images common in surveillance.
Existing methods often fail to accurately match LR probe images with high-resolution (HR) gallery images.

Purpose of the Study:

To develop a novel approach for matching LR probe images with HR gallery images.
To improve face recognition performance when input images are of significantly reduced resolution.

Main Methods:

Utilizing Multidimensional Scaling (MDS) to embed both LR and HR images into a common feature space.
Simultaneously learning mappings for LR and HR images using an iterative majorization algorithm.
Training the model on HR images to approximate HR-to-HR distances from LR-to-HR comparisons.

Related Experiment Videos

Last Updated: May 26, 2026

Quantification of Orofacial Phenotypes in Xenopus

Quantification of Orofacial Phenotypes in Xenopus

Published on: November 6, 2014

Main Results:

The proposed MDS-based method significantly improves face matching accuracy compared to direct LR matching or super-resolution pre-processing.
Demonstrated effectiveness on the Multi-PIE dataset with extremely low-resolution probes (down to 8x6 pixels).
Validated on real-world low-resolution surveillance images from the Surveillance Cameras Face Database.

Conclusions:

The novel MDS approach offers a robust solution for face recognition challenges posed by low-resolution imagery.
This method enhances the utility of surveillance systems and other applications relying on distant or degraded facial data.
Simultaneous embedding in a common space effectively bridges the resolution gap between probe and gallery images.