Abstract
In this paper, our objective is to propose a model binarization method aimed at addressing the challenges posed by over-parameterized super-resolution (SR) models. Our analysis reveals that binary SR models experience significant performance degradation, primarily attributed to their sensitivity towards weight/activation distributions, particularly when devoid of Batch Normalization (BN) layers. Consequently, we undertake the following endeavors in this study: First, we conduct a comprehensive analysis to examine the impact of BN layers on SR models based on Binary Neural Networks (BNNs). Second, we propose an asymmetric binarizer that can be reparameterized to adaptively adjust the transition point for activation binarization. Third, we introduce a progressive gradient estimator that modifies weight smoothness and controls weight flipping to stabilize the training procedure in the absence of BN layers. Through extensive experiments, we demonstrate that our proposed method exhibits significant performance improvements. For instance, when binarizing EDSR and scaling up input images by a factor of \(\times 4\), our approach achieves a PSNR decrease of less than 0.4dB on the Urban100 benchmark.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
BN layers are inserted within each channel. For simplicity, we omit the channel-wise index here.
References
Bevilacqua, M., Roumy, A., Guillemot, C., Morel, M.L.A.: Low-complexity single-image super-resolution based on nonnegative neighbor embedding. In: British Machine Vision Conference (BMVC) (2012)
Chen, B., et al.: Arm: any-time super-resolution method. In: Computer Vision-ECCV 2022: 17th European Conference, Tel Aviv, Israel, 23–27 October 2022, Proceedings, Part XIX, pp. 254–270. Springer (2022). https://doi.org/10.1007/978-3-031-19800-7
Chen, T., Zhang, Z., Ouyang, X., Liu, Z., Shen, Z., Wang, Z.: “bnn-bn=?”: training binary neural networks without batch normalization. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition (CVPR), pp. 4619–4629 (2021)
Dong, C., Loy, C.C., He, K., Tang, X.: Image super-resolution using deep convolutional networks. IEEE Trans. Pattern Anal. Mach. Intell. (TPAMI) 38(2), 295–307 (2015)
Dong, C., Loy, C.C., Tang, X.: Accelerating the super-resolution convolutional neural network. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9906, pp. 391–407. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46475-6_25
Gao, G., Li, W., Li, J., Wu, F., Lu, H., Yu, Y.: Feature distillation interaction weighting network for lightweight image super-resolution. In: Proceedings of the AAAI conference on artificial intelligence (AAAI), vol. 36, pp. 661–669 (2022)
He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778 (2016)
Helwegen, K., Widdicombe, J., Geiger, L., Liu, Z., Cheng, K.T., Nusselder, R.: Latent weights do not exist: Rethinking binarized neural network optimization. In: Advances in Neural Information Processing Systems (NeurIPS) 32 (2019)
Hong, C., Baik, S., Kim, H., Nah, S., Lee, K.M.: Cadyq: content-aware dynamic quantization for image super-resolution. In: Computer Vision-ECCV 2022: 17th European Conference, Tel Aviv, Israel, 23–27 October 2022, Proceedings, Part VII, pp. 367–383. Springer (2022). https://doi.org/10.1007/978-3-031-20071-7_22
Hong, C., Kim, H., Baik, S., Oh, J., Lee, K.M.: Daq: channel-wise distribution-aware quantization for deep image super-resolution networks. In: Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV), pp. 2675–2684 (2022)
Huang, G., Liu, Z., Van Der Maaten, L., Weinberger, K.Q.: Densely connected convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4700–4708 (2017)
Huang, J.B., Singh, A., Ahuja, N.: Single image super-resolution from transformed self-exemplars. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 5197–5206 (2015)
Jiang, X., Wang, N., Xin, J., Li, K., Yang, X., Gao, X.: Training binary neural network without batch normalization for image super-resolution. In: Proceedings of the AAAI Conference on Artificial Intelligence (AAAI), pp. 1700–1707 (2021)
Kim, J., Lee, J.K., Lee, K.M.: Accurate image super-resolution using very deep convolutional networks. In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition (CVPR), pp. 1646–1654 (2016)
Ledig, C., et al.: Photo-realistic single image super-resolution using a generative adversarial network. In: Proceedings of the IEEE conference on Computer Vision and Pattern Recognition (CVPR), pp. 4681–4690 (2017)
Ledig, C., et al.: Photo-realistic single image super-resolution using a generative adversarial network. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 4681–4690 (2017)
Li, H., et al.: PAMS: quantized super-resolution via parameterized max scale. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12370, pp. 564–580. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58595-2_34
Li, K., et al.: Local means binary networks for image super-resolution. IEEE Trans. Neural Netw. Learn. Syst. (TNNLS) (2022)
Lim, B., Son, S., Kim, H., Nah, S., Mu Lee, K.: Enhanced deep residual networks for single image super-resolution. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp. 136–144 (2017)
Lin, M., et al.: Rotated binary neural network. Adv. Neural Inform. Process. Syst. (NeurIPS) 33, 7474–7485 (2020)
Liu, Z., Shen, Z., Li, S., Helwegen, K., Huang, D., Cheng, K.T.: How do adam and training strategies help bnns optimization. In: International Conference on Machine Learning (ICML), pp. 6936–6946. PMLR (2021)
Liu, Z., et al.: Bi-Real Net: enhancing the performance of 1-Bit CNNs with improved representational capability and advanced training algorithm. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11219, pp. 747–763. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01267-0_44
Ma, Y., Xiong, H., Hu, Z., Ma, L.: Efficient super resolution using binarized neural network. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), (2019)
Martin, D., Fowlkes, C., Tal, D., Malik, J.: A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 416–423 (2001)
Qin, H., et al.: Forward and backward information retention for accurate binary neural networks. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition (CVPR), pp. 2250–2259 (2020)
Rastegari, M., Ordonez, V., Redmon, J., Farhadi, A.: XNOR-Net: imagenet classification using binary convolutional neural networks. In: Leibe, B., Matas, J., Sebe, N., Welling, M. (eds.) ECCV 2016. LNCS, vol. 9908, pp. 525–542. Springer, Cham (2016). https://doi.org/10.1007/978-3-319-46493-0_32
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Timofte, R., De Smet, V., Van Gool, L.: Anchored neighborhood regression for fast example-based super-resolution. In: Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 1920–1927 (2013)
Tipping, M., Bishop, C.: Bayesian image super-resolution. In: Advances in Neural Information Processing Systems (NeurIPS) 15 (2002)
Tu, Z., Chen, X., Ren, P., Wang, Y.: Adabin: improving binary neural networks with adaptive binary sets. In: Computer Vision-ECCV 2022: 17th European Conference, Tel Aviv, Israel, 23–27 October 2022, Proceedings, Part XI. pp. 379–395. Springer (2022). https://doi.org/10.1007/978-3-031-20083-0_23
Wang, P., He, X., Cheng, J.: Toward accurate binarized neural networks with sparsity for mobile application. IEEE Trans. Neural Netw. Learn. Syst. (TNNLS) (2022)
Wang, Z., Wu, Z., Lu, J., Zhou, J.: Bidet: an efficient binarized object detector. In: Proceedings of the IEEE/CVF conference on Computer Vision and Pattern Recognition (CVPR), pp. 2049–2058 (2020)
Xin, J., Wang, N., Jiang, X., Li, J., Huang, H., Gao, X.: Binarized neural network for single image super resolution. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12349, pp. 91–107. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58548-8_6
Xu, S., et al.: Ida-det: an information discrepancy-aware distillation for 1-bit detectors. In: Computer Vision-ECCV 2022: 17th European Conference, Tel Aviv, Israel, 23–27 October 2022, Proceedings, Part XI, pp. 346–361. Springer (2022). https://doi.org/10.1007/978-3-031-20083-0_21
Xu, S., Zhao, J., Lu, J., Zhang, B., Han, S., Doermann, D.: Layer-wise searching for 1-bit detectors. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 5682–5691 (2021)
Xu, Z., et al.: Recu: reviving the dead weights in binary neural networks. In: Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), pp. 5198–5208 (2021)
Zhang, Y., Tian, Y., Kong, Y., Zhong, B., Fu, Y.: Residual dense network for image super-resolution. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2472–2481 (2018)
Zhong, Y., et al.: Dynamic dual trainable bounds for ultra-low precision super-resolution networks. In: Computer Vision-ECCV 2022: 17th European Conference, Tel Aviv, Israel, 23–27 October 2022, Proceedings, Part XVIII, pp. 1–18. Springer (2022). https://doi.org/10.1007/978-3-031-19797-0_1
Zhu, Y., Zhang, Y., Yuille, A.L.: Single image super-resolution using deformable patches. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2917–2924 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2024 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Li, X., Chao, F. (2024). Binarizing Super-Resolution Neural Network Without Batch Normalization. In: Liu, Q., et al. Pattern Recognition and Computer Vision. PRCV 2023. Lecture Notes in Computer Science, vol 14434. Springer, Singapore. https://doi.org/10.1007/978-981-99-8549-4_6
Download citation
DOI: https://doi.org/10.1007/978-981-99-8549-4_6
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-8548-7
Online ISBN: 978-981-99-8549-4
eBook Packages: Computer ScienceComputer Science (R0)