Abstract
In this work, we propose a novel principal component approximation network (PCANet) for image compression. The proposed network is based on the assumption that a set of images can be decomposed into several shared feature matrices, and an image can be reconstructed by the weighted sum of these matrices. The proposed PCANet is specifically devised to learn and approximate these feature matrices and weight vectors, which are used to encode images for compression. Unlike previous deep learning-based methods, a distinctive aspect of our approach is its consideration of network size in the bit-rate computation. Despite this inclusion, our proposed method yields promising results. Through extensive experiments conducted on standard datasets, we demonstrate the effectiveness of our approach in comparison to state-of-the-art techniques. To the best of our knowledge, this is the first machine learning approach that includes the size of networks during bitrate computation in image compression.
- [1] . 2017. Soft-to-hard vector quantization for end-to-end learning compressible representations. Adv. Neural Inf. Process. Syst. 30 (2017).Google Scholar
- [2] . 2019. Generative Adversarial Networks for Extreme Learned Image Compression. (
Aug. 2019).DOI: Google ScholarCross Ref - [3] . 2018. Lossless image compression using reversible integer wavelet transforms and convolutional neural networks. In Proceedings of the Data Compression Conference. 395–395.
DOI: Google ScholarCross Ref - [4] . 1974. Discrete cosine transform. IEEE Trans. Comput. 100, 1 (1974), 90–93.Google ScholarDigital Library
- [5] . 2022. Efficient light field image compression with enhanced random access. ACM Trans. Multim. Comput., Commun. Applic. 18, 2 (
Mar. 2022), 44:1–44:18.DOI: Google ScholarDigital Library - [6] . 1968. Fourier transform coding of images. In Proceedings of the Hawaii International Conference: System Sciences. 677–679.Google Scholar
- [7] . 2017. Learning to inpaint for image compression. In Advances in Neural Information Processing Systems, , , , , , , and (Eds.), Vol. 30. Curran Associates, Inc.Retrieved from https://proceedings.neurips.cc/paper/2017/file/013a006f03dbc5392effeb8f18fda755-Paper.pdfGoogle Scholar
- [8] . 2016. End-to-end optimized image compression. In Proceedings of the 5th International Conference on Learning Representations.Google Scholar
- [9] . 2018. Variational image compression with a scale hyperprior. In Proceedings of the 6th International Conference on Learning Representations.Google Scholar
- [10] . 2001. Calculation of average PSNR differences between RD-curves. ITU SG16 Doc. VCEG-M33 (2001).Google Scholar
- [11] . 2021. End-to-end learnt image compression via non-local attention optimization and improved context modeling. IEEE Trans. Image Process. 30 (2021), 3179–3191.
DOI: Google ScholarCross Ref - [12] . 2020. Variable bitrate image compression with quality scaling factors. In Proceedings of the IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP’20). 2163–2167.
DOI: Google ScholarCross Ref - [13] . 2019. Learning image and video compression through spatial-temporal energy compaction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’19).Google ScholarCross Ref
- [14] . 2020. Learned image compression with discretized Gaussian mixture likelihoods and attention modules. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’20).Google ScholarCross Ref
- [15] . 2000. The JPEG2000 still image coding system: An overview. IEEE Trans. Consum. Electron. 46, 4 (2000), 1103–1127.Google ScholarDigital Library
- [16] 2014. CDnet 2014: An expanded change detection benchmark dataset. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW’14). 393–400.
DOI: Google ScholarDigital Library - [17] . 2021. Neural image compression via attentional multi-scale back projection and frequency decomposition. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV’21). 14677–14686.Google ScholarCross Ref
- [18] . 2023. OpenDMC: An open-source library and performance evaluation for deep-learning-based multi-frame compression. In Proceedings of the ACM International Conference on Multimedia (ACM MM’23).Google ScholarDigital Library
- [19] . 2016. Towards conceptual compression. In Advances in Neural Information Processing Systems, , , , , and (Eds.), Vol. 29. Curran Associates, Inc.Google Scholar
- [20] . 2015. DRAW: A recurrent neural network for image generation. In Proceedings of the 32nd International Conference on Machine Learning (Proceedings of Machine Learning Research), and (Eds.), Vol. 37. PMLR, 1462–1471.Google Scholar
- [21] . 2022. ELIC: Efficient learned image compression with unevenly grouped space-channel contextual adaptive coding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 5718–5727.Google ScholarCross Ref
- [22] . 2022. Learning end-to-end lossy image compression: A benchmark. IEEE Trans. Pattern Anal. Mach. Intell. 44, 8 (2022), 4194–4211.
DOI: Google ScholarDigital Library - [23] . 1952. A Method for the Construction of Minimum-Redundancy Codes. Proc. IRE 40, 9 (
Sept. 1952), 1098–1101.DOI: Google ScholarCross Ref - [24] . 2018. Improved lossy image compression with priming and spatially adaptive bit rates for recurrent networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18).Google ScholarCross Ref
- [25] . 2022. Joint global and local hierarchical priors for learned image compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 5992–6001.Google ScholarCross Ref
- [26] . 2019. Context-adaptive entropy model for end-to-end optimized image compression. In Proceedings of the 7th International Conference on Learning Representations.Google Scholar
- [27] . 2022. DPICT: Deep progressive image compression using trit-planes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 16113–16122.Google ScholarCross Ref
- [28] . 2022. Deep stereo image compression via bi-directional coding. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 19669–19678.Google ScholarCross Ref
- [29] . 2018. Learning convolutional networks for content-weighted image compression. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18).Google ScholarCross Ref
- [30] . 2014. Microsoft COCO: Common objects in context. In Proceedings of the European Conference on Computer Vision. Springer, 740–755.Google ScholarCross Ref
- [31] Ziwei Liu, Ping Luo, Xiaogang Wang, and Xiaoou Tang. 2015. Deep learning face attributes in the wild. In Proceedings of International Conference on Computer Vision (ICCV’15).Google Scholar
- [32] . 2019. Image and video compression with neural networks: A review. IEEE Trans. Circ. Syst. Vid. Technol. 30, 6 (2019), 1683–1698.Google ScholarCross Ref
- [33] . 2018. Conditional probability models for deep image compression. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’18).Google ScholarCross Ref
- [34] . 2018. Joint autoregressive and hierarchical priors for learned image compression. In Advances in Neural Information Processing Systems, , , , , , and (Eds.), Vol. 31. Curran Associates, Inc.Retrieved from https://proceedings.neurips.cc/paper/2018/file/53edebc543333dfbf7c5933af792c9c4-Paper.pdfGoogle Scholar
- [35] . 2021. Neural Compression. Retrieved from https://github.com/facebookresearch/NeuralCompre ssionGoogle Scholar
- [36] . 2021. Saliency driven perceptual image compression. In Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 227–236.Google ScholarCross Ref
- [37] . 1969. Hadamard transform image coding. Proc. IEEE 57, 1 (1969), 58–68.Google ScholarCross Ref
- [38] . 2022. LC-FDNet: Learned lossless image compression with frequency decomposition network. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 6033–6042.Google ScholarCross Ref
- [39] . 2017. Real-time adaptive image compression. In Proceedings of the 34th International Conference on Machine Learning (Proceedings of Machine Learning Research), and (Eds.), Vol. 70. PMLR, 2922–2930.Google Scholar
- [40] . 2015. On robust image spam filtering via comprehensive visual modeling. Pattern Recog. 48, 10 (
Oct. 2015), 3227–3238.DOI: Google ScholarDigital Library - [41] . 2021. BBAS: Towards large scale effective ensemble adversarial attacks against deep neural network learning. Inf. Sci. 569 (
Aug. 2021), 469–478.DOI: Google ScholarCross Ref - [42] . 2021. Variable-rate deep image compression through spatially-adaptive feature transform. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV’21). 2380–2389.Google ScholarCross Ref
- [43] . 2017. Lossy image compression with compressive autoencoders. In Proceedings of the 5th International Conference on Learning Representations.Google Scholar
- [44] . 2017. Full resolution image compression with recurrent neural networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’17).Google ScholarCross Ref
- [45] . 1992. The JPEG still picture compression standard. IEEE Trans. Consum. Electron. 38, 1 (
Feb. 1992), xviii–xxxiv.DOI: Google ScholarDigital Library - [46] . 2022. Neural data-dependent transform for learned image compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 17379–17388.Google ScholarCross Ref
- [47] . 2003. Multiscale structural similarity for image quality assessment. In Proceedings of the 37th Asilomar Conference on Signals, Systems & Computers, Vol. 2. IEEE, 1398–1402.Google ScholarCross Ref
- [48] . 1987. Arithmetic coding for data compression. Commun. ACM 30, 6 (
June 1987), 520–540.DOI: Google ScholarDigital Library - [49] . 2022. SASIC: Stereo image compression with latent shifts and stereo attention. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 661–670.Google ScholarCross Ref
- [50] . 2020. A GAN-based tunable image compression system. In Proceedings of the IEEE Winter Conference on Applications of Computer Vision (WACV’20). 2323–2331.
DOI: Google ScholarCross Ref - [51] . 2018. Variational autoencoder for low bit-rate image compression. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2617–2620.Google Scholar
- [52] . 2022. Unified multivariate gaussian mixture for efficient neural image compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 17612–17621.Google ScholarCross Ref
- [53] . 2022. The devil is in the details: Window-based attention for image compression. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR’22). 17492–17501.Google ScholarCross Ref
Index Terms
- Principal Component Approximation Network for Image Compression
Recommendations
Conditional Entropy Coding of VQ Indexes for Image Compression
DCC '97: Proceedings of the Conference on Data CompressionVector quantization (VQ) is a source coding methodology with provable rate-distortion optimality. However, despite more than two decades of intensive research, VQ theoretical promise is yet to be fully realized in image compression practice. Restricted ...
Switching of Wavelet Transforms by Neural Network for Image Compression
Nowadays, digital images compression requires more and more significant attention of researchers. Even when high data rates are available, image compression is necessary in order to reduce the memory used, as well the transmission cost. An ideal image ...
Neuro-Wavelet Based Approach for Image Compression
CGIV '07: Proceedings of the Computer Graphics, Imaging and VisualisationImages have large data quantity. For storage and transmission of images, high efficiency image compression methods are under wide attention. In this paper we propose a neuro- wavelet based model for image compression which combines the advantage of ...
Comments