检索结果-内蒙古大学图书馆

learned image compression With Efficient Cross-Platform Entropy Coding

IEEE JOURNAL ON EMERGING AND SELECTED TOPICS IN CIRCUITS AND SYSTEMS 2025年第1期15卷 72-82页

作者： Yang, Runyu Liu, Dong Wu, Feng Gao, Wen Univ Sci & Technol China Key Lab Brain Inspired Intelligent Percept & Cogni MOE Hefei 230027 Peoples R China Peking Univ Sch Comp Sci Beijing 100871 Peoples R China Peng Cheng Lab Shenzhen 518055 Peoples R China

learned image compression has shown remarkable compression efficiency gain over the traditional image compression solutions, which is partially attributed to the learned entropy models and the adopted entropy coding engine. However, the inference of the entropy models and the sequential nature of the entropy coding both incur high time complexity. Meanwhile, the neural network-based entropy models usually involve floating-point computations, which incur inconsistent probability estimation and decoding failure in different platforms. We address these limitations by introducing an efficient and cross-platform entropy coding method, chain coding-based latent compression (CC-LC), into learned image compression. First, we leverage the classic chain coding and carefully design a block-based entropy coding procedure, significantly reducing the number of coding symbols and thus the coding time. Second, since CC-LC is not based on neural networks, we propose a rate estimation network as a surrogate of CC-LC during the end-to-end training. Third, we alternately train the analysis/synthesis networks and the rate estimation network for the rate-distortion optimization, making the learned latent fit CC-LC. Experimental results show that our method achieves much lower time complexity than the other learned image compression methods, ensures cross-platform consistency, and has comparable compression efficiency with BPG. Our code and models are publicly available at https://***/Yang-Runyu/CC-LC.

关键词： image coding Estimation Entropy coding Entropy Computational modeling Training Symbols Transforms Circuits and systems Time complexity Block-based coding chain coding cross-platform consistency entropy coding learned image compression

来源：评论

学校读者我要写书评

暂无评论

Q-LIC: Quantizing learned image compression With Channel Splitting

引用

IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY 2025年第4期35卷 3798-3811页

作者： Sun, Heming Yu, Lu Katto, Jiro Waseda Univ Waseda Res Inst Sci & Engn Tokyo 1698555 Japan Japan Sci & Technol Agcy JST PRESTO Kawaguchi Saitama 3320012 Japan Zhejiang Univ Inst Informat & Commun Engn Hangzhou 310023 Peoples R China Waseda Univ Grad Sch Fundamental Sci & Engn Tokyo 1698555 Japan Waseda Res Inst Sci & Engn Tokyo 1698555 Japan

learned image compression (LIC) has reached a comparable coding gain with traditional hand-crafted methods such as VVC intra. However, the large network complexity prohibits the usage of LIC on resource-limited embedded systems. Network quantization is an efficient way to reduce the network burden. This paper presents a quantized LIC (QLIC) by channel splitting. First, we explore that the influence of quantization error to the reconstruction error is different for various channels. Second, we split the channels whose quantization has larger influence to the reconstruction error. After the splitting, the dynamic range of channels is reduced so that the quantization error can be reduced. Finally, we prune several channels to keep the number of overall channels as origin. By using the proposal, in the case of 8-bit quantization for weight and activation of both main and hyper path, we can reduce the BD-rate by 0.61%-4.74% compared with the previous QLIC. Besides, we can reach better coding gain compared with the state-of-the-art network quantization method when quantizing MS-SSIM models. Moreover, our proposal can be combined with other network quantization methods to further improve the coding gain. The moderate coding loss caused by the quantization validates the feasibility of the hardware implementation for QLIC in the future.

关键词： Quantization (signal) image coding Transforms Mathematical models image reconstruction Arithmetic Hardware learned image compression quantization fixed-point channel splitting

来源：评论

学校读者我要写书评

暂无评论

STanH: Parametric Quantization for Variable Rate learned image compression

引用

IEEE TRANSACTIONS ON image PROCESSING 2025年 34卷 639-651页

作者： Presta, Alberto Tartaglione, Enzo Fiandrotti, Attilio Grangetto, Marco Univ Turin Comp Sci Dept I-10124 Turin Italy Inst Polytech Paris LTCI Telecom Paris F-91120 Palaiseau France

In end-to-end learned image compression, encoder and decoder are jointly trained to minimize a R + lambda D cost function, where lambda controls the trade-off between rate of the quantized latent representation and image quality. Unfortunately, a distinct encoder-decoder pair with millions of parameters must be trained for each lambda , hence the need to switch encoders and to store multiple encoders and decoders on the user device for every target rate. This paper proposes to exploit a differentiable quantizer designed around a parametric sum of hyperbolic tangents, called STanH, that relaxes the step-wise quantization function. STanH is implemented as a differentiable activation layer with learnable quantization parameters that can be plugged into a pre-trained fixed rate model and refined to achieve different target bitrates. Experimental results show that our method enables variable rate coding with comparable efficiency to the state-of-the-art, yet with significant savings in terms of ease of deployment, training time, and storage costs.

关键词： Quantization (signal) image coding Training Adaptation models Decoding Bit rate Noise Entropy Context modeling Annealing learned image compression variable rate image coding differentiable quantization quantizer annealing

来源：评论

学校读者我要写书评

暂无评论

Generalized Gaussian Model for learned image compression

引用

IEEE TRANSACTIONS ON image PROCESSING 2025年 34卷 1950-1965页

作者： Zhang, Haotian Li, Li Liu, Dong Univ Sci & Technol China MOE Key Lab Brain Inspired Intelligent Percept & C Hefei 230093 Peoples R China

In learned image compression, probabilistic models play an essential role in characterizing the distribution of latent variables. The Gaussian model with mean and scale parameters has been widely used for its simplicity and effectiveness. Probabilistic models with more parameters, such as the Gaussian mixture models, can fit the distribution of latent variables more precisely, but the corresponding complexity is higher. To balance the compression performance and complexity, we extend the Gaussian model to the generalized Gaussian family for more flexible latent distribution modeling, introducing only one additional shape parameter beta than the Gaussian model. To enhance the performance of the generalized Gaussian model by alleviating the train-test mismatch, we propose improved training methods, including beta -dependent lower bounds for scale parameters and gradient rectification. Our proposed generalized Gaussian model, coupled with the improved training methods, is demonstrated to outperform the Gaussian and Gaussian mixture models on a variety of learned image compression networks.

关键词： image coding Training Probabilistic logic Transforms Entropy Complexity theory Quantization (signal) Shape Table lookup Noise Generalized Gaussian model learned image compression probabilistic model

来源：评论

学校读者我要写书评

暂无评论

Subset-Selection Weight Post-Training Quantization Method for learned image compression Task

引用

IEEE ACCESS 2025年 13卷 5145-5153页

作者： Yang, Jinru Wang, Xiaoqin Li, Qiang Qiao, Shushan Zhou, Yumei Chinese Acad Sci Inst Microelect Beijing 100029 Peoples R China Univ Chinese Acad Sci Sch Integrated Circuits Beijing 101408 Peoples R China

Post-training quantization(PTQ) has been widely studied in recent years because it does not require retraining the network or the entire training dataset. However, naively applying the PTQ method to quantize low-level image tasks such as learned image compression(LIC) usually incurs significant accuracy degradation. Existing solutions often aim to optimize the calibration process or minimize quantization loss in the context of uniform quantization, which makes it difficult to further reduce the quantization loss. This work achieves accurate weight quantization using a non-uniform quantization method called subset-selection within the PTQ method. Subset-selection method uses a clustering algorithm to select a proper batch of quantization points from a large uniformly constructed quantization points pool that matches the current layer's weight distribution correctly. Compared to the state-of-the-art LIC PTQ method, experimental results show that our proposed method achieves an average 0.8% better BD-rate on several LIC models. Meanwhile, we extended our method to image classification models and achieved an average 0.64% better accuracy, further proving the generalization of our method.

关键词： Quantization (signal) image coding Calibration Clustering algorithms Decoding Accuracy Linear programming image reconstruction Computational modeling Adaptation models Convolutional neural network post-training quantization non-uniform quantization learned image compression

来源：评论

学校读者我要写书评

暂无评论

JND-LIC: learned image compression via Just Noticeable Difference for Human Visual Perception

引用

IEEE TRANSACTIONS ON BROADCASTING 2025年第1期71卷 217-228页

作者： Pan, Zhaoqing Zhang, Guoyu Peng, Bo Lei, Jianjun Xie, Haoran Wang, Fu Lee Ling, Nam Tianjin Univ Sch Elect & Informat Engn Tianjin 300072 Peoples R China Lingnan Univ Dept Comp & Decis Sci Hong Kong Peoples R China Hong Kong Metropolitan Univ Sch Sci & Technol Hong Kong Peoples R China Santa Clara Univ Dept Comp Sci & Engn Santa Clara CA 95053 USA

Existing human visual perception-oriented image compression methods well maintain the perceptual quality of compressed images, but they may introduce fake details into the compressed images, and cannot dynamically improve the perceptual rate-distortion performance at the pixel level. To address these issues, a just noticeable difference (JND)-based learned image compression (JND-LIC) method is proposed for human visual perception in this paper, in which a weight-shared model is used to extract image features and JND features, and the learned JND features are utilized as perceptual prior knowledge to assist the image coding process. In order to generate a highly compact image feature representation, a JND-based feature transform module is proposed to model the pixel-to-pixel masking correlation between the image features and the JND features. Furthermore, inspired by eye movement research that the human visual system perceives image degradation unevenly, a JND-guided quantization mechanism is proposed for the entropy coding, which adjusts the quantization step of each pixel to further eliminate perceptual redundancies. Extensive experimental results show that our proposed JND-LIC significantly improves the perceptual quality of compressed images with fewer coding bits compared to state-of-the-art learned image compression methods. Additionally, the proposed method can be flexibly integrated with various advanced learned image compression methods, and has robust generalization capabilities to improve the efficiency of perceptual coding.

关键词： image compression learned image compression just noticeable difference transform quantization

来源：评论

学校读者我要写书评

暂无评论

Dynamic kernel-based adaptive spatial aggregation for learned image compression

引用

JOURNAL OF VISUAL COMMUNICATION AND image REPRESENTATION 2025年 110卷

作者： Wang, Huairui Fu, Nianxiang Chen, Zhenzhong Liu, Shan Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan Peoples R China Wuhan Univ Hubei Luojia Lab Wuhan Peoples R China Tencent Amer Media Lab Palo Alto CA USA

learned image compression methods have shown remarkable performance and expansion potential compared to traditional codecs. Currently, there are two mainstream image compression frameworks: one uses stacked convolution and other uses window-based self-attention for transform coding, most of which aggregate valuable dependencies in a fixed spatial range. In this paper, we focus on extending content-adaptive aggregation capability and propose a dynamic kernel-based transform coding. The proposed adaptive aggregation generates kernel offsets to capture valuable information with dynamic sampling convolution to help transform. With the adaptive aggregation strategy and the sharing weights mechanism, our method can achieve promising transform capability with acceptable model complexity. Besides, considering the coarse hyper prior, the channel-wise, and the spatial context, we formulate a generalized entropy model. Based on it, we introduce dynamic kernel in hyper-prior to generate more expressive side information context. Furthermore, we propose an asymmetric sparse entropy model according to the investigation of the spatial and variance characteristics of the grouped latents. The proposed entropy model can facilitate entropy coding to reduce statistical redundancy while maintaining inference efficiency. Experimental results demonstrate that our method achieves superior rate-distortion performance on three benchmarks compared to the state-of-the-art learning-based methods.

关键词： learned image compression Adaptive spatial aggregation Asymmetric sparse entropy model

来源：评论

学校读者我要写书评

暂无评论

Enhancing learned image compression via Cross Window-Based Attention 19th

Enhancing Learned Image Compression via Cross Window-Based A...

引用

19th International Symposium on Visual Computing

作者： Mudgal, Priyanka Liu, Feng Portland State Univ Portland OR 97124 USA

ISBN: (纸本)9783031773884;9783031773891

In recent years, learned image compression methods have demonstrated superior rate-distortion performance compared to traditional image compression methods. Recent methods utilize convolutional neural networks (CNN), variational autoencoders (VAE), invertible neural networks (INN), and transformers. Despite their significant contributions, a main drawback of these models is their poor performance in capturing local redundancy. Therefore, to leverage global features along with local redundancy, we propose a CNN-based solution integrated with a feature encoding module. The feature encoding module encodes important features before feeding them to the CNN and then utilizes cross-scale window-based attention, which further captures local redundancy. Crossscale window-based attention is inspired by the attention mechanism in transformers and effectively enlarges the receptive field. Both the feature encoding module and the cross-scale window-based attention module in our architecture are flexible and can be incorporated into any other network architecture. We evaluate our method on the Kodak and CLIC datasets and demonstrate that our approach is effective and on par with state-of-the-art methods.

关键词： learned image compression end-to-end image compression

来源：评论

学校读者我要写书评

暂无评论

ALIC: Adaptive Fusion Entropy Model for learned image compression

ALIC: Adaptive Fusion Entropy Model for Learned Image Compre...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Li, Lingxue Liu, Meiqin Zhang, Yifan Tang, Qi Yao, Chao Zhao, Yao Institute of Information Science Beijing Jiaotong University Beijing China Visual Intelligence + X International Cooperation Joint Laboratory of MOE Beijing Jiaotong University China School of Computer and Communication Engineering University of Science and Technology Beijing Beijing China

ISBN: (纸本)9798350368741

Recently, learned image compression algorithms have achieved significant performance. The entropy model is crucial for improving the rate-distortion performance by estimating the probability distribution of latent representation. In this paper, we propose an adaptive fusion entropy model for learned image compression (ALIC). To explore the correlation between channel and global spatial features, an adaptive fusion entropy model (AFEM) is designed. AFEM first slices the latent representation along the channels and leverages the adaptive channel fusion context module (ACFC) to capture correlations between the decoded and current slices. Subsequently, AFEM uses the adaptive spatial fusion context module (ASFC) to further divide the current slice into encoding pixels and reference pixels, thus improving the accuracy of probability estimation. The attention map and modulation parameter are introduced in ACFC and ASFC to interact with channel and spatial features. In addition, the variable-rate residual transformer (VResFormer) is proposed to control dynamic bit-rate by selectively modulating the high-frequency component according to coefficient weight and bias. Experimental results indicate that our ALIC outperforms other learned image compression algorithms. Our ALIC saves 5.89% bit-rate compared with VVC (4:4:4) on Kodak dataset. © 2025 IEEE.

关键词： Entropy model learned image compression Variable-rate

来源：评论

学校读者我要写书评

暂无评论

Estimating the resize parameter in end-to-end learned image compression

引用

SIGNAL PROCESSING-image COMMUNICATION 2025年 135卷

作者： Chen, Li-Heng Bampis, Christos G. Li, Zhi Krasula, Lukas Bovik, Alan C. Univ Texas Austin Dept Elect & Comp Engn Austin TX 78712 USA Netflix Inc Los Gatos CA USA

We describe a search-free resizing framework that can further improve the rate-distortion tradeoff of recent learned image compression models. Our approach is simple: compose a pair of differentiable downsampling/upsampling layers that sandwich a neural compression model. To determine resize factors for different inputs, we utilize another neural network jointly trained with the compression model, with the end goal of minimizing the rate-distortion objective. Our results suggest that "compression friendly"downsampled representations can be quickly determined during encoding by using an auxiliary network and differentiable image warping. By conducting extensive experimental tests on existing deep image compression models, we show results that our new resizing parameter estimation framework can provide Bj & oslash;ntegaard-Delta rate (BD- rate) improvement of about 10% against leading perceptual quality engines. We also carried out a subjective quality study, the results of which show that our new approach yields favorable compressed images. To facilitate reproducible research in this direction, the implementation used in this paper is being made freely available online at: https://***/treammm/Resizecompression.

关键词： Resizing Convolutional neural networks learned image compression

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：