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 Videos

Optimizing multiscale SSIM for compression via MLDS.

Christophe Charrier¹, Kenneth Knoblauch, Laurence T Maloney

¹Université de Caen-Basse Normandie, GREYC Laboratory, Equipe Image, ENSICAEN, Caen 14050, France. christophe.charrier@unicaen.fr

IEEE Transactions on Image Processing : a Publication of the IEEE Signal Processing Society

|August 8, 2012

Summary

This summary is machine-generated.

Related Concept Videos

Methods of Medium Optimization

Methods of Medium Optimization

Optimizing growth media enhances microbial proliferation and maximizes product yield. Statistical experimental design methodologies provide structured and reproducible approaches, offering progressively higher levels of robustness and efficiency.The One-Factor-at-a-Time (OFAT) MethodThe One-Factor-at-a-Time (OFAT) method involves adjusting a single variable while keeping all others constant. However, it cannot detect interactions between variables, often leading to suboptimal outcomes when...

Upsampling

Upsampling

Managing signal sampling rates is essential in digital signal processing to maintain signal integrity. A decimated signal, characterized by a reduced frequency range due to its lower sampling rate, can be upsampled by inserting zeros between each sample. This upsampling process expands the original spectrum and introduces repeated spectral replicas at intervals dictated by the new Nyquist frequency. To refine this zero-inserted sequence, it is passed through a lowpass filter with a cutoff...

You might also read

Related Articles

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

Sort by

Same author

Effects of color-enhancing filters on color salience in normal trichromats.

Vision research·2026

Same author

Joint Quality Assessment and Example-Guided Tone Mapping by Disentangling Picture Appearance From Content.

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

Same author

Quality Prediction of Embedded and Overlaid Text in User-Generated Visual Content.

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

Same author

SAMScore: A Content Structural Similarity Metric for Image Translation Evaluation.

IEEE transactions on artificial intelligence·2026

Same author

HoloQA: Full Reference Video Quality Assessor of Rendered Human Avatars in Virtual Reality.

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

Same author

Effects of color-enhancing filters on color salience in normal trichromats.

bioRxiv : the preprint server for biology·2025

Same journal

Hyperbolic Cycle Alignment for Infrared-Visible Image Fusion.

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

Same journal

Learning Gaze Synthesizer via 3D-eye Controlled Diffusion and Cross-domain Feature Alignment.

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

Same journal

Underlying Semantic Diffusion for Effective and Efficient In-Context Learning.

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

Same journal

DiffRES: Unleashing Text-to-Image Diffusion Models for Generative Referring Expression Segmentation without Information Leakage.

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

Same journal

Location Matters: Frequency-Spatial Dual Space Adaptation for Cross-Domain Few-Shot Segmentation.

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

Same journal

BayeTopo: Bayesian-based Topology-guided Learning for Vascular Imaging Segmentation.

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

See all related articles

Maximum Likelihood Difference Scaling (MLDS) offers a robust method for evaluating image compression quality. This approach refines image quality assessment algorithms like MS-SSIM by quantifying perceptual differences more accurately.

Area of Science:

Computer Vision
Image Processing
Perceptual Quality Assessment

Background:

Assessing perceived quality of compressed images is vital for evaluating compression methods.
Traditional methods rely on subjective observer ratings, which have limitations like inter-observer variability and overlooking specific artifacts.
Existing image quality assessment algorithms, such as multiscale structural similarity (MS-SSIM), are often calibrated using these subjective ratings.

Purpose of the Study:

To introduce and apply Maximum Likelihood Difference Scaling (MLDS) for a more objective assessment of perceived image quality.
To evaluate the performance of the MS-SSIM algorithm using the MLDS method.
To demonstrate how MLDS data can be used to recalibrate and improve the performance of image quality assessment algorithms.

Related Experiment Videos

Main Methods:

Utilized Maximum Likelihood Difference Scaling (MLDS) to quantify supra-threshold perceptual differences between image pairs.
Applied MLDS to assess perceived image quality across a wide range of images and compression rates.
Analyzed the performance of the multiscale structural similarity (MS-SSIM) algorithm in conjunction with MLDS data.

Main Results:

MLDS provides a quantitative measure of perceptual differences, overcoming subjectivity in traditional rating scales.
The study examined how perceived image quality, as estimated by MLDS, varies with increasing compression levels.
Data derived from MLDS enabled the recalibration of the MS-SSIM algorithm, leading to improved performance.

Conclusions:

MLDS is a powerful tool for assessing perceived image quality and understanding compression artifact perception.
The MLDS approach circumvents the limitations of traditional subjective rating methods.
Recalibrating image quality assessment algorithms like MS-SSIM using MLDS data enhances their accuracy in predicting human perception.