BANet: Blur-aware Attention Networks for Dynamic Scene Deblurring
This addresses the problem of non-uniform motion blur in images for applications like photography and video processing, offering an incremental improvement in efficiency and accuracy.
The paper tackles dynamic scene deblurring by proposing BANet, which achieves accurate and efficient deblurring via a single forward pass, performing favorably against state-of-the-art methods on benchmarks like GoPro and RealBlur with real-time results.
Image motion blur results from a combination of object motions and camera shakes, and such blurring effect is generally directional and non-uniform. Previous research attempted to solve non-uniform blurs using self-recurrent multiscale, multi-patch, or multi-temporal architectures with self-attention to obtain decent results. However, using self-recurrent frameworks typically lead to a longer inference time, while inter-pixel or inter-channel self-attention may cause excessive memory usage. This paper proposes a Blur-aware Attention Network (BANet), that accomplishes accurate and efficient deblurring via a single forward pass. Our BANet utilizes region-based self-attention with multi-kernel strip pooling to disentangle blur patterns of different magnitudes and orientations and cascaded parallel dilated convolution to aggregate multi-scale content features. Extensive experimental results on the GoPro and RealBlur benchmarks demonstrate that the proposed BANet performs favorably against the state-of-the-arts in blurred image restoration and can provide deblurred results in real-time.