Improving Multi-generation Robustness of Learned Image Compression
Not on the map. Why?
Abstract (arXiv)
Benefit from flexible network designs and end-to-end joint optimization approach, learned image compression (LIC) has demonstrated excellent coding performance and practical feasibility in recent years. However, existing compression models suffer from serious multi-generation loss, which always occurs during image editing and transcoding. During the process of repeatedly encoding and decoding, the quality of the image will rapidly degrade, resulting in various types of distortion, which significantly limits the practical application of LIC. In this paper, a thorough analysis is carried out to determine the source of generative loss in successive image compression (SIC). We point out and solve the quantization drift problem that affects SIC, reversibility loss function as well as channel relaxation method are proposed to further reduce the generation loss. Experiments show that by using our proposed solutions, LIC can achieve comparable performance to the first compression of BPG even after 50 times reencoding without any change of the network structure.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 21 of its 44 references that are in the map. First posted on arXiv 31 Oct 2022.
Builds on
21 of 44 references.
References · 21
- 2017Attention Is All You Needcited by 59,932
- 2015Deep Residual Learning for Image Recognitioncited by 51,809
- 2013Auto-Encoding Variational Bayescited by 3,980
- 2018Variational image compression with a scale hyperpriorcited by 910
- 2018Joint Autoregressive and Hierarchical Priors for Learned Image Compressioncited by 642
- 2020Learned Image Compression with Discretized Gaussian Mixture Likelihoods and Attention Modulescited by 570
- 2022ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Codingcited by 296
- 2017Lossy Image Compression with Compressive Autoencoderscited by 289
- 2020CompressAI: a PyTorch library and evaluation platform for end-to-end compression researchcited by 271
- 2020Channel-wise Autoregressive Entropy Models for Learned Image Compressioncited by 255
- 2018Context-adaptive Entropy Model for End-to-end Optimized Image Compressioncited by 138
- 2020Nonlinear Transform Codingcited by 132
9 more
- 2021Enhanced Invertible Encoding for Learned Image Compressioncited by 102
- 2020Asymmetric Gained Deep Image Compression With Continuous Rate Adaptationcited by 83
- 2020Causal Contextual Prediction for Learned Image Compressioncited by 70
- 2020A Unified End-to-End Framework for Efficient Deep Image Compressioncited by 56
- 2021Soft then Hard: Rethinking the Quantization in Neural Image Compressioncited by 50
- 2020Lossy Image Compression with Normalizing Flowscited by 25
- 2021Towards Robust Neural Image Compression: Adversarial Attack and Model Finetuningcited by 24
- 2022Towards Bidirectional Arbitrary Image Rescaling: Joint Optimization and Cycle Idempotencecited by 18
- 2021Pseudocylindrical Convolutions for Learned Omnidirectional Image Compressioncited by 6
Built on by
Citing papers per year: 2023: 1.
1 paper.