Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Hybrid-Domain High-Frequency Attention Network for Arbitrary Magnification Super-Resolution

임의배율 초해상도를 위한 하이브리드 도메인 고주파 집중 네트워크

  • Yun, Jun-Seok (Department of AI Convergence, Chonnam National University) ;
  • Lee, Sung-Jin (Department of AI Convergence, Chonnam National University) ;
  • Yoo, Seok Bong (Department of AI Convergence, Chonnam National University) ;
  • Han, Seunghwoi (School of Mechanical Engineering, Chonnam National University)
  • Received : 2021.08.18
  • Accepted : 2021.09.10
  • Published : 2021.11.30
Download PDF
⟨ Previous Next ⟩

Abstract

Recently, super-resolution has been intensively studied only on upscaling models with integer magnification. However, the need to expand arbitrary magnification is emerging in representative application fields of actual super-resolution, such as object recognition and display image quality improvement. In this paper, we propose a model that can support arbitrary magnification by using the weights of the existing integer magnification model. This model converts super-resolution results into the DCT spectral domain to expand the space for arbitrary magnification. To reduce the loss of high-frequency information in the image caused by the expansion by the DCT spectral domain, we propose a high-frequency attention network for arbitrary magnification so that this model can properly restore high-frequency spectral information. To recover high-frequency information properly, the proposed network utilizes channel attention layers. This layer can learn correlations between RGB channels, and it can deepen the model through residual structures.

최근 이미지 초해상도는 정수배율만 가능한 모델에만 집중적으로 연구되고 있다. 하지만 관심 객체 인식, 디스플레이 화질 개선 등 실제 초해상도 기술의 대표 적용 분야에서는 소수 배율을 포함하는 임의배율 확대 필요성이 대두되고 있다. 본 논문에서는 기존 정수배율 모델의 가중치를 활용하여 임의배율을 실행할 수 있는 모델을 제안한다. 이 모델은 정수배율에 의해 우수한 성능을 가진 초해상도 결과를 DCT 스펙트럼 도메인으로 변환하여 임의배율을 위한 공간을 확장한다. DCT 스펙트럼 도메인에 의한 확장으로 인해 발생하는 이미지의 고주파 정보 손실 문제를 줄이기 위해 고주파 스펙트럼 정보를 적절히 복원할 수 있는 모델인 고주파 집중 네트워크를 제안한다. 제안된 네트워크는 고주파 정보를 제대로 생성하기 위해서 RGB 채널간의 상관관계를 학습하는 레이어인 channel attention을 활용하고, 잔차 학습 구조를 통해 모델을 깊게 만들어 성능을 향상시켰다.

Keywords

Acknowledgement

This work was supported by the National Research Foundation of Korea(NRF) grant funded by the Korea government(MSIT)(NRF-2020R1A4A1019191) and Institute of Information & Communications Technology Planning & Evaluation(IITP) grant funded by the Korea government(MSIT) (No.2020-0-00004, Development of Previsional Intelligence based on Long-term Visual Memory Network).

References

  1. S. J. Lee, T. J. Kim, C. H. Lee, and S. B. Yoo, "Image super-resolution for improving object recognition accuracy," in Journal of the Korea Institute of Information and Communication Engineering, vol. 25, no. 6, pp. 774-784, Jun. 2021. https://doi.org/10.6109/JKIICE.2021.25.6.774
  2. C. Dong, C. C. Loy, K. He, and X. Tang, "Image super-resolution using deep convolutional networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 38, no. 2, pp. 295-307, Feb. 2016. https://doi.org/10.1109/TPAMI.2015.2439281
  3. J. W. Kim, J. K. Lee, and K. M. Lee, "Accurate image super-resolution using very deep convolutional networks," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1646-1654, 2016.
  4. W. Shi, J. Caballero, F. Huszar, J. Totz, A. P. Aitken, R. Bishop, D. Rueckert, and Z. Wang, "Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network," in Proceeding of the European Conference on Computer Vision, pp. 1874-1883, Sep. 2016.
  5. M. Haris, G. Shakhnarovich, and N. Ukita, "Deep back-projection networks for super-resolution," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1664-1673, Mar. 2018.
  6. Y. Zhang, K. Li, K. Li, L. Wang, B. Zhong, and Y. Fu, "Image super-resolution using very deep residual channel attention networks," in Proceeding of the European Conference on Computer Vision, pp. 286-301, Mar. 2018.
  7. Y. Guo, J. Chen, J. Wang, Q. Chen, J. Cao, Z. Deng, Y. Xu, and M. Tan, "Closed-loop matters: Dual regression networks for single image super-resolution," in Proceeding of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 5407-5416, May. 2020.
  8. Y. Zhang, Y. Tian, Y. Kong, B. Zhong, and Y. Fu, "Residual dense network for image super-resolution," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2472-2481, Mar. 2018.
  9. T. Dai, J. Cai, Y. Zhang, S. T. Xia, and L. Zhang, "Second-order attention network for single image super-resolution," in Proceeding of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 11065-11074, 2019.
  10. C. Ledig, L. Theis, F. Huszar, J. Caballero, A. Cunningham, A. Acosta, A. Aitken, A. Tejani, J. Totz, and Z. Wang, "Photo-realistic single image super-resolution using a generative adversarial network," in Proceeding of the European Conference on Computer Vision, pp. 4681-4690, 2017.
  11. A. Lugmayr, M. Danelljan, L. V. Gool, and R. Timofte, "Srflow: Learning the super-resolution space with normalizing flow," in Proceeding of the European Conference on Computer Vision, pp. 715-732, 2020.
  12. Y. H. Jo, S. J. Yang, and S. J. Kim, "Srflow-da: Super-resolution using normalizing flow with deep convolutional block," in Proceeding of the European Conference on Computer Vision Workshops, pp. 364-372, June. 2021.
  13. Y. G. Kim and D. H. Son, "Noise conditional flow model for learning the super-resolution," in Proceeding of the European Conference on Computer Vision Workshops, pp. 424-432, June. 2021.
  14. W. Shi, J. Caballero, F. Huszar, J. Totz, A. P. Aitken, R. Bishop, D. Rueckert, and Z. Wang, "Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network," in Proceeding of the European Conference on Computer Vision, pp. 1874-1883, 2016.
  15. X. Hu, H. Mu, X. Zhang, Z. Wang, T. Tan, and J. Sun, "Meta-SR: A magnification-arbitrary network for super-resolution," in Proceeding of the European Conference on Computer Vision, pp. 1575-1584, 2019.
  16. S. H. Son and K. M. Lee, "SRWarp: Generalized image super-resolution under arbitrary transformation," in Proceeding of the European Conference on Computer Vision, pp. 7782-7791, June. 2021.
  17. N. Kumar, R. Verma, and A. Sethi, "Convolutional neural networks for wavelet domain super resolution," in Pattern Recognition Letters, vol. 90, pp. 65-71, 2017. https://doi.org/10.1016/j.patrec.2017.03.014
  18. J. Li, S. You, and A. Robles-Kelly, "A frequency domain neural network for fast image super-resolution," in International Joint Conference on Neural Networks, pp. 1-8, 2018.
  19. S. Xue, W. Qiu, F. Liu, and X. Jin, "Faster image super-resolution by improved frequency-domain neural networks," Signal, Image and Video Processing, vol. 14, no. 2, pp. 257-265, 2020. https://doi.org/10.1007/s11760-019-01548-8
  20. O. Aydin and R. G. Cinbis, "Single-image super-resolution analysis in DCT spectral domain," in Balkan Journal of Electrical & Computer Engineering, vol. 8, no. 3, pp. 209-217, 2020.
  21. Z. Liu, P. Luo, X. Wang, and X. Tang, "Large-scale CelebFaces Attributes (CelebA) Dataset," in Retrieved August, vol. 2018.15, no. 11, 2018.
  22. R. Timofte, E. Agustsson, L. Van Gool, M.-H. Yang, L. Zhang, B. Lim, S. Son, H. Kim, S. Nah, and K. M. Lee, "NTIRE 2017 challenge on single image super-resolution: Methods and results," in Proceeding of the European Conference on Computer Vision Workshops, July. 2017.
  23. D. Martin, C. Fowlkes, D. Tal, and J. Malik, "A database of human segmented natural images and its application to evaluating segmentation algorithms and measuring ecological statistics," in International Conference on Computer Vision, pp. 416-423, 2001.