DehazeSB: Real-World Dehazing via Schrödinger Bridge (ICCV 2025) by Liu ChangDehazeSB: Real-World Dehazing via Schrödinger Bridge (ICCV 2025) by Liu Chang

DehazeSB: Real-World Dehazing via Schrödinger Bridge (ICCV 2025)

Portfolio project
Liu Chang
Motivation & Challenge: Real-world dehazing needs flexible distribution mapping without sacrificing image details. DehazeSB formulates dehazing as a Schrödinger Bridge that transports hazy images directly to clear ones, with detail-preserving regularization and learned haze-aware CLIP guidance. Results & Venues: Improved real-world dehazing through direct hazy-to-clear transport in fewer sampling steps. Published at ICCV 2025. Link: https://arxiv.org/pdf/2507.09524

1 like · 0 views · Sep 15, 2026

Motivation & Challenge: Real-world dehazing needs flexible distribution mapping without sacrificing image details.

Tags