Adaptive High-Pass Kernel Prediction for Efficient Video Deblurring

TL;DR

This paper introduces an efficient video deblurring method that explicitly captures high-frequency details using adaptive high-pass kernels, achieving state-of-the-art results with low memory and fast inference.

Contribution

It proposes a novel adaptive high-pass kernel prediction approach that enhances high-frequency detail recovery in video deblurring, addressing neural network spectral bias.

Findings

Achieves state-of-the-art performance among low-budget models.

Low-memory footprint and fast inference times.

Effective high-frequency detail extraction in blurred videos.

Abstract

State-of-the-art video deblurring methods use deep network architectures to recover sharpened video frames. Blurring especially degrades high-frequency (HF) information, yet this aspect is often overlooked by recent models that focus more on enhancing architectural design. Recovering these fine details is challenging, partly due to the spectral bias of neural networks, which are inclined towards learning low-frequency functions. To address this, we enforce explicit network structures to capture the fine details and edges. We dynamically predict adaptive high-pass kernels from a linear combination of high-pass basis kernels to extract high-frequency features. This strategy is highly efficient, resulting in low-memory footprints for training and fast run times for inference, all while achieving state-of-the-art when compared to low-budget models. The code is available at https://github.com/jibo27/AHFNet.