Special Session 10

Special Session 10: DIT-Driven Visual Enhancement: Restoration, Super-Resolution & Beyond

Description: This special session focuses on recent advances in image restoration, enhancement, and super-resolution leveraging Diffusion-based Image Transformers (DIT). DIT models have shown remarkable potential in generating high-fidelity visual outputs by effectively combining diffusion probabilistic models with transformer architectures. The session aims to bring together researchers working on DIT-based methods for low-level vision tasks, such as denoising, deblurring, inpainting, resolution upscaling, and perceptual quality improvement. We will explore how DIT frameworks can be adapted, optimized, and extended to address real-world degradation scenarios while maintaining computational efficiency and generalization across diverse datasets.

Session organizer
Assoc. Prof. Tingting Yang, University of Sanya, China

The topics of interest include, but are not limited to:
• Diffusion models for image restoration
• Transformer-based architectures in low-level vision
• Image super-resolution and quality enhancement
• Degradation-aware generative modeling
• Efficient inference and lightweight DIT design
• Multi-modal conditioning for image reconstruction
• Evaluation metrics and benchmarks for DIT-enhanced images

Submission method
Submit your Full Paper (no less than 8 pages) or your paper abstract-without publication (200-400 words) via Online Submission System, then choose Special Session 10 (DIT-Driven Visual Enhancement: Restoration, Super-Resolution & Beyond)
Template Download

Introduction of session organizers