ASF-Net: Robust Video Deraining via Temporal Alignment and Online Adaptive Learning
This work improves video deraining for applications like surveillance and autonomous driving, but it is incremental as it builds on existing learning-based methods with specific enhancements.
The paper tackles video deraining by addressing challenges in exploiting temporal correlations and adapting to real-world scenarios, proposing ASF-Net with a temporal shift module and a re-degraded learning strategy, which achieves superior performance on three benchmarks and compelling visual quality in real-world scenes.
In recent times, learning-based methods for video deraining have demonstrated commendable results. However, there are two critical challenges that these methods are yet to address: exploiting temporal correlations among adjacent frames and ensuring adaptability to unknown real-world scenarios. To overcome these challenges, we explore video deraining from a paradigm design perspective to learning strategy construction. Specifically, we propose a new computational paradigm, Alignment-Shift-Fusion Network (ASF-Net), which incorporates a temporal shift module. This module is novel to this field and provides deeper exploration of temporal information by facilitating the exchange of channel-level information within the feature space. To fully discharge the model's characterization capability, we further construct a LArge-scale RAiny video dataset (LARA) which also supports the development of this community. On the basis of the newly-constructed dataset, we explore the parameters learning process by developing an innovative re-degraded learning strategy. This strategy bridges the gap between synthetic and real-world scenes, resulting in stronger scene adaptability. Our proposed approach exhibits superior performance in three benchmarks and compelling visual quality in real-world scenarios, underscoring its efficacy. The code is available at https://github.com/vis-opt-group/ASF-Net.