Over-fitting-based image compression requires weights compactness for compression and fast convergence for practical use, posing challenges for deep convolutional neural networks (CNNs) based methods. This paper presents a simple re-parameterization method to train CNNs with reduced weights storage and accelerated convergence. The convolution kernels are re-parameterized as a weighted sum of discrete cosine transform (DCT) kernels enabling direct optimization in the frequency domain. Combined with L1 regularization, the proposed method surpasses vanilla convolutions by achieving a significantly improved rate-distortion with low computational cost. The proposed method is verified with extensive experiments of over-fitting-based image restoration on various datasets, achieving up to -46.12% BD-rate on top of HEIF with only 200 iterations.
@article{arxiv.2310.08068,
title = {Frequency-Aware Re-Parameterization for Over-Fitting Based Image Compression},
author = {Yun Ye and Yanjie Pan and Qually Jiang and Ming Lu and Xiaoran Fang and Beryl Xu},
journal= {arXiv preprint arXiv:2310.08068},
year = {2023}
}
Comments
to be published at ICIP 2023, this version fixed a mistake in Eq. (1) in the proceeding version