Unified Removal of Raindrops and Reflections: A New Benchmark and A Novel Pipeline

Abstract

When capturing images through glass surfaces or windshields on rainy days, raindrops and reflections frequently co-occur to significantly reduce the visibility of captured images. This practical problem lacks attention and needs to be resolved urgently. Prior de-raindrop, de-reflection, and all-in-one models have failed to address this composite degradation. To this end, we first formally define the unified removal of raindrops and reflections (UR$^3$) task for the first time and construct a real-shot dataset, namely RainDrop and ReFlection (RDRF), which provides a new benchmark with substantial, high-quality, diverse image pairs. Then, we propose a novel diffusion-based framework (i.e., DiffUR$^3$) with several target designs to address this challenging task. By leveraging the powerful generative prior, DiffUR$^3$ successfully removes both types of degradations. Extensive experiments demonstrate that our method achieves state-of-the-art performance on our benchmark and on challenging in-the-wild images. The RDRF dataset and the codes will be made public upon acceptance.

Figures (22)

• Figure 1: We compare our DiffUR$^3$ pipeline with other methods on low-quality images with raindrops and reflections from our newly collected real-world benchmark. Specifically, (c) DAI Hu2026AAAI-DAI is designed for reflection removal, (d) A cascaded method, and (e) A re-trained all-in-one method (i.e., Histoformer Sun2024ECCV-Hist and $\dagger$ indicates re-trained on our dataset), (f) Our DiffUR$^3$ pipeline jointly removes both degradations in a single pass.
• Figure 2: (a) Sketch diagram and actual equipment of our image acquisition platform. To suppress shutter-induced micro-vibrations which may potentially induce image misalignment, we implement a wireless triggering mechanism. It comprises a remote controller and a camera-mounted signal receiver, enabling contact-free shutter operation. (b) The data collection pipeline for our RDRF dataset. denotes light occlusion
• Figure 3: Our RDRF dataset comprises a diverse collection of scenes, each contains a ground truth and multiple low-quality images. As illustrated in this figure, the clean ground truths are highlighted in red boxes, while corresponding low-quality images are arranged around. We divide it into the training and testing subsets, ensuring no overlapping samples between them. Please zoom in on screen for a better view.
• Figure 4: (a) Overall pipeline of our DiffUR$^3$ framework. Given a low-quality image $I_{lq}$, the restoration stage removes the undesired degradation to obtain the initial result $I_{s}$. Both $I_{lq}$ and $I_{s}$ are fed into the next stage as the condition images. We inject the effective condition information through a control branch, which outputs control signals for the noise prediction U-Net. (b) Details of the Modulate&Gate module within the control branch. (c) The generation of noisy latent $z_{t}$ during the training phase. Note that the noisy latent starts from random Gaussian noise during the inference.
• Figure 5: The motivation of employing an additional fidelity encoder.