Frequency-Aware Perceptual Optimization for Low-Complexity Implicit Image Compression

Haotian Wu, Gen Li, Di You, Pier Luigi Dragotti, Deniz Gunduz
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:136283-136330, 2026.

Abstract

We propose a frequency-aware perceptual optimization framework for low-complexity image compression, realized as a Realism-enhanced Region-based Implicit Codec (Re2IC). Re2IC models visual perception via saliency-guided region partitioning and local–global perceptual modulation. To enhance realism under complexity constraints, we introduce wavelet–Wasserstein distortion (WA-WD), a frequency-decomposed perceptual distortion that balances fidelity and realism through subband-aware modeling and provides a more reliable approximation than standard Wasserstein distortion. Together, these designs enable fine-grained spatial–spectral optimization, allowing Re2IC to achieve superior rate–perception trade-offs, outperforming generative codecs such as HiFiC while using less than 1% of their decoding cost. Extensive experiments demonstrate state-of-the-art perceptual performance among overfitted codecs. Beyond compression, WA-WD serves as a standalone, tunable perceptual metric with strong alignment to human preference (Pearson 94.6%, Spearman 92.3%) and competitive performance across multiple IQA benchmarks. Project page: https://eedavidwu.github.io/ReReIC/

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-wu26ae, title = {Frequency-Aware Perceptual Optimization for Low-Complexity Implicit Image Compression}, author = {Wu, Haotian and Li, Gen and You, Di and Dragotti, Pier Luigi and Gunduz, Deniz}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {136283--136330}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/wu26ae/wu26ae.pdf}, url = {https://proceedings.mlr.press/v306/wu26ae.html}, abstract = {We propose a frequency-aware perceptual optimization framework for low-complexity image compression, realized as a Realism-enhanced Region-based Implicit Codec (Re2IC). Re2IC models visual perception via saliency-guided region partitioning and local–global perceptual modulation. To enhance realism under complexity constraints, we introduce wavelet–Wasserstein distortion (WA-WD), a frequency-decomposed perceptual distortion that balances fidelity and realism through subband-aware modeling and provides a more reliable approximation than standard Wasserstein distortion. Together, these designs enable fine-grained spatial–spectral optimization, allowing Re2IC to achieve superior rate–perception trade-offs, outperforming generative codecs such as HiFiC while using less than 1% of their decoding cost. Extensive experiments demonstrate state-of-the-art perceptual performance among overfitted codecs. Beyond compression, WA-WD serves as a standalone, tunable perceptual metric with strong alignment to human preference (Pearson 94.6%, Spearman 92.3%) and competitive performance across multiple IQA benchmarks. Project page: https://eedavidwu.github.io/ReReIC/} }
Endnote
%0 Conference Paper %T Frequency-Aware Perceptual Optimization for Low-Complexity Implicit Image Compression %A Haotian Wu %A Gen Li %A Di You %A Pier Luigi Dragotti %A Deniz Gunduz %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-wu26ae %I PMLR %P 136283--136330 %U https://proceedings.mlr.press/v306/wu26ae.html %V 306 %X We propose a frequency-aware perceptual optimization framework for low-complexity image compression, realized as a Realism-enhanced Region-based Implicit Codec (Re2IC). Re2IC models visual perception via saliency-guided region partitioning and local–global perceptual modulation. To enhance realism under complexity constraints, we introduce wavelet–Wasserstein distortion (WA-WD), a frequency-decomposed perceptual distortion that balances fidelity and realism through subband-aware modeling and provides a more reliable approximation than standard Wasserstein distortion. Together, these designs enable fine-grained spatial–spectral optimization, allowing Re2IC to achieve superior rate–perception trade-offs, outperforming generative codecs such as HiFiC while using less than 1% of their decoding cost. Extensive experiments demonstrate state-of-the-art perceptual performance among overfitted codecs. Beyond compression, WA-WD serves as a standalone, tunable perceptual metric with strong alignment to human preference (Pearson 94.6%, Spearman 92.3%) and competitive performance across multiple IQA benchmarks. Project page: https://eedavidwu.github.io/ReReIC/
APA
Wu, H., Li, G., You, D., Dragotti, P.L. & Gunduz, D.. (2026). Frequency-Aware Perceptual Optimization for Low-Complexity Implicit Image Compression. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:136283-136330 Available from https://proceedings.mlr.press/v306/wu26ae.html.

Related Material