MID: A Self-supervised Multimodal Iterative Denoising Framework

TL;DR

MID is a novel self-supervised framework that iteratively denoises complex non-linear noisy data across multiple domains without needing paired datasets, achieving state-of-the-art results.

Contribution

The paper introduces a self-supervised multimodal iterative denoising framework that models noise as a continuous process and learns to remove it without paired clean-noisy data.

Findings

Achieves state-of-the-art performance on four computer vision tasks.

Demonstrates robustness and adaptability in biomedical and bioinformatics applications.

Effectively models complex non-linear noise through iterative learning.

Abstract

Data denoising is a persistent challenge across scientific and engineering domains. Real-world data is frequently corrupted by complex, non-linear noise, rendering traditional rule-based denoising methods inadequate. To overcome these obstacles, we propose a novel self-supervised multimodal iterative denoising (MID) framework. MID models the collected noisy data as a state within a continuous process of non-linear noise accumulation. By iteratively introducing further noise, MID learns two neural networks: one to estimate the current noise step and another to predict and subtract the corresponding noise increment. For complex non-linear contamination, MID employs a first-order Taylor expansion to locally linearize the noise process, enabling effective iterative removal. Crucially, MID does not require paired clean-noisy datasets, as it learns noise characteristics directly from the noisy inputs. Experiments across four classic computer vision tasks demonstrate MID's robustness, adaptability, and consistent state-of-the-art performance. Moreover, MID exhibits strong performance and adaptability in tasks within the biomedical and bioinformatics domains.