PixelBoost 8 – Pixel Quality with 8X Highlights Boosting Enhancement
Article Sidebar
Main Article Content
Abstract
In recent years, deep learning has become a fundamental technology across a wide array of scientific and industrial fields, largely fuelled by advances in computational capabilities. One area that has experienced substantial progress is face hallucination—the task of improving the resolution of facial images. This process is critical to various computer vision applications, including facial recognition, feature extraction, and identity verification. Recently, deep generative models, particularly Generative Adversarial Networks (GANs), have led the field. Although these models have produced remarkable results, there is still a pressing need to further improve both accuracy and output quality. In order to address these problems, we propose a new GAN-based face hallucination method. This method is primarily based on the Enhanced Super-Resolution Generative Adversarial Network (ESRGAN). We present a personalised adaptation of ESRGAN that employs the VGG16 architecture with a compact pre-trained version. This method balances output image quality and computational efficiency. Experiments show that our approach is effective. The improved model obtains a maximum peak signal-to-noise ratio (PSNR) of 30.30. The Learned Perceptual Image Patch Similarity (LPIPS) score is 0.0817, whereas the Structural Similarity Index Measure (SSIM) is 0.8757. The results surpass many state-of-the-art methods available today. These enhancements have a significant impact and importance.
Downloads
Article Details
Section
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
How to Cite
References
(2022). Low-resolution face recognition based on feature-mapping face hallucination. Computers and Electrical Engineering 101.
DOI: 10.1016/j.compeleceng.2022.108136
Aakerberg, A., Nasrollahi, K. & Moeslund, T. B. (2021). Real-world super-resolution of face images from surveillance cameras. IET Image Processing 16(8), pp. 1234-1245. DOI: 10.1049/ipr2.12359
Grm, K., Scheirer, W. J. & Struc, V. (2020). Face Hallucination Using Cascaded Super-Resolution and Identity Priors. IEEE Transactions on Image Processing 29(1), pp. 2150-2165. DOI: 10.1109/TIP.2019.2945835
Wang, Z., She, Q. & Ward, T. E. (2021). Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy. ACM Computing Surveys 54(2). DOI: 10.1145/3439723
Tustison, N. J., Avants, B. B. & Gee, J. C. (2019). Learning image-based spatial transformations via convolutional neural networks: A review. Magnetic Resonance Imaging 64, pp. 142-153. DOI: 10.1016/j.mri.2019.05.037
(2023). ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks - Technical Documentation and Model Overview. ESRGAN ReadTheDocs. https://esrgan.readthedocs.io/en/latest/index.html
Zhu, Z., Lei, Y., Qin, Y. & Zhu, C. (2023). IRE: Improved Image Super-Resolution Based on Real-ESRGAN. IEEE Access 99, p. 1.
DOI: 10.1109/ACCESS.2023.3256086
Choi, Y. & Park, H. (2023). Improving ESRGAN with an additional image quality loss. Multimedia Tools & Applications 82(2), p. 3123. DOI: 10.1007/s11042-022-13452-4
Tomar, A. S., Arya, K. & Rajput, S. S. (2024). Learning face super-resolution through identity features and distilling facial prior knowledge. Expert Systems with Applications 202. DOI: 10.1016/j.eswa.2024.125625
Bashir, S. M., Wang, Y., Khan, M. & Niu, Y. (2021). A comprehensive review of deep learning-based single image super-resolution. PeerJ Computer Science 7. DOI: 10.7717/peerj-cs.621
Zhang, W., Zhao, W., Li, J., Zhuang, P., Sun, H., Xu, Y. & Li, C. (2024). CVANet: Cascaded Visual Attention Network for Single
Image Super-Resolution. Neural Networks 170, pp. 622-634. DOI: 10.1016/j.neunet.2023.11.049
Zhu, Y., Zhang, Y. & Yuille, A. L. (2014). Single Image Super-Resolution Using Deformable Patches. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition, pp. 2917-2924. DOI: 10.1109/CVPR.2014.373, works remain significant, see the declaration
Anwar, S., Khan, S. & Barnes, N. (2020). A Deep Journey into Super-resolution: A Survey. ACM Computing Surveys 53(3).
DOI: 10.1145/3390462
Yu, X., Fernando, B., Hartley, R. & Porikli, F. (2020). Semantic Face Hallucination: Super-Resolving Very Low-Resolution Face Images with Supplementary Attributes. IEEE Transactions on Pattern Analysis and Machine Intelligence 42(11), pp. 2926-2943.
DOI: 10.1109/TPAMI.2019.2916881
Rana, M. S., Nibali, A. & He, Z. (2022). Selection of object detections using overlap map predictions. Neural Computing and Applications 34. DOI: 10.1007/s00521-022-07469-x
Hu, X., Fan, Z., Jia, X., Li, Z., Zhang, X., Qi, L. & Xuan, Z. (2021). Towards effective learning for face super-resolution with shape and pose perturbations. Knowledge-Based Systems 220. DOI: 10.1016/j.knosys.2021.106938
Dastmalchi, H. & Aghaeinia, H. (2022). Super-resolution of very low-resolution face images with a wavelet-integrated, identity-preserving adversarial network. Image and Vision Computing 107. DOI: 10.1016/j.image.2022.116755
(2020). Facial Image Synthesis and Super-Resolution with Stacked Generative Adversarial Network. Neurocomputing 402, pp. 359-365. DOI: 10.1016/j.neucom.2020.03.107
Arabboev, M., Begmatov, S., Rikhsivoev, M., Nosirov, K. & Saydiakbarov, S. (2021). A comprehensive review of image super-resolution metrics: classical and AI-based approaches. Acta IMEKO 13(1). DOI: 10.21014/actaimeko.v13i1.1679
(2021). Py-Feat: Python Facial Expression Analysis Toolbox. PMC10751270. DOI: 10.3389/fnins.2021.635019
He, K., Pu, N., Lao, M. & Lew, M. S. (2023). Few-shot and meta-learning methods for image understanding: a survey. International Journal of Multimedia Information Retrieval 12. DOI: 10.1007/s13735-023-00279-4
Lopes, A., Santos, F. P., Oliveira, D. d., Schiezaro, M. & Pedrini, H. (2024). Computer Vision Model Compression Techniques for Embedded Systems: A Survey. Computers & Graphics 123. DOI: 10.1016/j.cag.2024.104015
Zhang, Z., Shi, Y., Zhou, X., Kan, H. & Wen, J. (2020). Shuffle block SRGAN for face image super-resolution reconstruction. Measurement and Control 53(78), pp. 1429-1439. DOI: 10.1177/0020294020944969
Mishra, A., Lee, B. & Lee, B. (2025). PixelBoost: Leveraging Brownian Motion for Realistic-Image Super-Resolution. IEEE Transactions on Multimedia 99, pp. 1-13. DOI: 10.1109/TMM.2025.3623517
Zhang, Y., Tian, Y., Kong, Y., Zhong, B. & Fu, Y. (2018). Residual Dense Network for Image Super-Resolution. arXiv preprint arXiv:1802.08797. DOI: 10.48550/arXiv.1802.08797
Li, M., Zhang, Z., Yu, J. & Chen, C. (2020). Learning Face Image Super-Resolution Through Facial Semantic Attribute Transformation and Self-Attentive Structure Enhancement. IEEE Transactions on Multimedia 99, p. 1. DOI: 10.1109/TMM.2020.2984092
Musunuri, Y. R. & Kwon, O. (2021). Deep Residual Dense Network for Single Image Super-Resolution. Electronics 10(5).
DOI: 10.3390/electronics10050555
Fan, Z., Hu, X., Chen, C., Wang, X. & Peng, S. (2020). Facial image super-resolution guided by adaptive geometric features. EURASIP Journal on Wireless Communications and Networking 2020. DOI: 10.1186/s13638-020-01760-y
(2021). Improved Face Image Super-Resolution Model Based on Generative Adversarial Network. MDPI 11(5).
DOI: 10.3390/23134331
(2022). Deep Learning for Face Super-Resolution: A Techniques Review. APSIPA Transactions on Signal and Information Processing 11(1), pp. 1-15. DOI: 10.1561/116.20240045
Johnson, J., Alahi, A. & Fei-Fei, L. (2016). Perceptual Losses for Real-Time Style Transfer and Super-Resolution. arXiv preprint arXiv:1603.08155. DOI: 10.1007/978-3-319-46475-6_43
Georgopoulos, M., Oldfield, J., Nicolaou, M. A., Panagakis, Y. & Pantic, M. (2021). Mitigating Demographic Bias in Facial Datasets with Style-Based Multi-attribute Transfer. International Journal of Computer Vision 129. DOI: 10.1007/s11263-021-01448-w
(2020). Facial image super-resolution guided by adaptive geometric features. Journal on Wireless Communications and Networking.
DOI: 10.1186/s13638-020-01760-y
Joze, H. R., Zharkov, I., Powell, K., Ringler, C., Liang, L., Roulston, A., Lutz, M. & Pradeep, V. (2020). ImagePairs: Realistic Super Resolution Dataset via Beam Splitter Camera Rig. IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops.
DOI: 10.1109/CVPRW50498.2020.00267
Tejy, A. R., Halder, S. S., Shandeelya, A. P. & Pankajakshany, V. (2020). Enhancing Perceptual Loss with Adversarial Feature Matching for Super-Resolution. arXiv preprint. https://arxiv.org/pdf/2005.07502
Liu, Y., Liu, Y. & Liu, Y. (2017). Tensor Super-Resolution with Generative Adversarial Nets. IEEE Transactions on Image Processing 26(1), pp. 1-12. DOI: 10.1109/TIP.2016.2610190
Liu, Y., Sun, D., Wang, F., Lim, K. P., Chiew, T. K. & Lai, Y. (2022). Improved Face Image Super-Resolution Model Based on Generative Adversarial Network. IEEE Transactions on Circuits and Systems for Video Technology 32(1), pp. 1-14.
DOI: 10.1109/TCSVT.2021.3071234
Kim, D., Kim, M., Kwon, G. & Kim, D. (2019). Progressive Face Super-Resolution via Attention to Facial Landmark. arXiv preprint arXiv:1908.08239. DOI: 10.48550/arXiv.1908.08239
eugenesiow. (2026). Dataset Card for Div2k. Hugging Face. https://huggingface.co/datasets/eugenesiow/Div2k
Wang, X., Yu, K., Wu, S., Gu, J., Liu, Y., Dong, C., Loy, C. C., Qiao, Y. & Tang, X. (2018). ESRGAN: Enhanced Super-Resolution Generative Adversarial Networks. ECCVW 2018. DOI: 10.1109/ICCVW.2018.00135
(2021). Synchronizing billion-scale automata. Information Sciences 574, pp. 162-175. DOI: 10.1016/j.ins.2021.05.072
(2024). Peak signal-to-noise ratio. Wikipedia. https://en.wikipedia.org/wiki/Peak_signal-to-noise_ratio
Karras, T., Aittala, M., Hellsten, J., Laine, S., Lehtinen, J. & Aila, T. (2020). Training Generative Adversarial Networks with Limited Data. arXiv preprint arXiv:2006.06676. DOI: 10.48550/arXiv.2006.06676
He, K., Zhang, X., Ren, S. & Sun, J. (2016). Deep Residual Learning for Image Recognition. IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770-778. DOI: 10.1109/CVPR.2016.90
(2024). Learned Perceptual Image Patch Similarity (LPIPS). PyTorch-Metrics 0.8.0 documentation. https://torchmetrics.readthedocs.io/en/v0.8.0/image/learned_perceptual_image_patch_similarity.html