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

Related Concept Videos

Downsampling

Downsampling

When considering a sampled sequence with zero values between sampling instants, one can replace it by taking every N-th value of the sequence. At these integer multiples of N, the original and sampled sequences coincide. This process, known as decimation, involves extracting every N-th sample from a sequence, thereby creating a more efficient sequence.
The Fourier transform of the decimated sequence reveals a combination of scaled and shifted versions of the original spectrum. This...

Reconstruction of Signal using Interpolation

Reconstruction of Signal using Interpolation

Signal processing techniques are essential for accurately converting continuous signals to digital formats and vice versa. When a continuous signal is sampled with a period T, the resulting sampled signal exhibits replicas of the original spectrum in the frequency domain, spaced at intervals equal to the sampling frequency. To handle this sampled signal, a zero-order hold method can be applied, which creates a piecewise constant signal by retaining each sample's value until the next sampling...

You might also read

Related Articles

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

Sort by

Same author

Spin radical enhanced magnetocapacitance effect in intermolecular excited states.

The journal of physical chemistry. B·2013

Same author

Recent developments in stir bar sorptive extraction.

Analytical and bioanalytical chemistry·2013

Same author

Discovery of MK-8742: an HCV NS5A inhibitor with broad genotype activity.

ChemMedChem·2013

Same author

Magnetic polycarbonate microspheres for tumor-targeted delivery of tumor necrosis factor.

Drug delivery·2013

Same author

A study on validity of cortical alpha connectivity for schizophrenia.

Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Annual International Conference·2013

Same author

Myosin light chain 2-based selection of human iPSC-derived early ventricular cardiac myocytes.

Stem cell research·2013

Search research articles

Related Experiment Video

Updated: May 31, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

Image restoration model compression via mamba-oriented heterogeneous knowledge distillation.

Sai Yang¹, Bin Hu², Xiaoxin Wu¹

¹School of Electrical Engineering and Automation, Nantong University, Nantong, 226019, China.

Neural Networks : the Official Journal of the International Neural Network Society

|May 29, 2026

Summary

We developed Mamba-oriented Heterogeneous Knowledge Distillation (MHKD) to compress image restoration (IR) models. This method efficiently transfers knowledge from large transformer models to smaller Mamba models, achieving high performance with fewer parameters.

Keywords:

Heterogeneous knowledge distillation Image restoration Model compression

Related Experiment Videos

Last Updated: May 31, 2026

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Constructing and Visualizing Models using Mime-based Machine-learning Framework

Published on: July 22, 2025

Area of Science:

Computer Vision
Artificial Intelligence
Machine Learning

Background:

Transformer-based image restoration (IR) models are computationally intensive and parameter-heavy.
There is a need for efficient IR models without compromising performance.

Purpose of the Study:

To propose a knowledge distillation strategy for compressing IR models.
To develop a Mamba-oriented Heterogeneous Knowledge Distillation (MHKD) framework for efficient IR model compression.

Main Methods:

Pre-trained a high-quality transformer-based IR model as the teacher network.
Constructed a lightweight Mamba-based IR model as the student network.
Employed MHKD with feature filter and feature interface modules for heterogeneous knowledge transfer.

Main Results:

MHKD successfully transferred knowledge from transformer to Mamba IR models.
The Mamba-based student model achieved comparable or superior performance to state-of-the-art models.
The compressed Mamba model has a compact size of approximately 716K parameters.

Conclusions:

MHKD is an effective strategy for compressing IR models.
Knowledge distillation enables efficient Mamba-based IR models.
The proposed method achieves a favorable balance between model size and performance.