Frequency-Aware Re-Parameterization for Over-Fitting Based Image Compression
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.
Code (0)
등록된 구현이 없습니다.
Tasks
Image CompressionImage RestorationMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
FiRe: Frequency Reparameterization as a Preconditioner for Periodic Implicit Neural Representations
Periodic Implicit Neural Representations (INRs) such as SIREN and FINER assign every neuron, the same global frequency, spending the representational budget inefficiently when local signal content varies. We introduce Fi…
FitDiT: Advancing the Authentic Garment Details for High-fidelity Virtual Try-on
Although image-based virtual try-on has made considerable progress, emerging approaches still encounter challenges in producing high-fidelity and robust fitting images across diverse scenarios. These methods often strugg…
Virtual Try-onBenign Overfitting in Classification: Provably Counter Label Noise with Larger Models
Studies on benign overfitting provide insights for the success of overparameterized deep learning models. In this work, we examine whether overfitting is truly benign in real-world classification tasks. We start with the…
FACT-GS: Frequency-Aligned Complexity-Aware Texture Reparameterization for 2D Gaussian Splatting
Realistic scene appearance modeling has advanced rapidly with Gaussian Splatting, which enables real-time, high-quality rendering. Recent advances introduced per-primitive textures that incorporate spatial color variatio…
Optimal Eye Surgeon: Finding Image Priors through Sparse Generators at Initialization
We introduce Optimal Eye Surgeon (OES), a framework for pruning and training deep image generator networks. Typically, untrained deep convolutional networks, which include image sampling operations, serve as effective im…
Image GenerationImage Restoration