Raw Image Based Over-Exposure Correction
Fig.1. Example scenes in our RPO dataset.
RAW image-based Real-world Paired Over-exposure dataset
Overexposed images present a unique set of challenges that can severely undermine image integrity. Overexposure can lead to washed-out regions, where important details are lost, colors appear faded or completely bleached, and the overall contrast of the image is compromised. This degradation can be particularly detrimental for applications relying on fine-grained visual cues, such as pattern recognition and texture analysis. In light of these issues, the advancement of overexposure correction techniques has become a crucial endeavor in the field of image processing, aiming to retrieve lost information in overexposed areas, rebalance color distributions, and reinstate the dynamic range that is crucial for maintaining visual fidelity. This work is integral to improving the robustness of computer vision systems when faced with images that have been excessively exposed to light.
To propel research in this field forward, it is essential to assess proposed methods in real-world scenarios. Consequently, we will utilize the RAW image-based Real-world Paired Over-exposure (RPO) dataset, introduced by Prof. Fu’s team in [a], captured using a Canon EOS 5D Mark IV camera. The RPO dataset comprises paired images collected across various scenes. Each short-exposure (normal-exposure) image is paired with long-exposure (over-exposure) images with 4 ratios (x3, x5, x8, x10).
We will host the competition using open source online platform, e.g. CodaLab. All submissions are evaluated by our script running on the server and we will double check the results of top-rank methods manually before releasing the final test-set rating.