ABSTRACT
Mirrors often appear in various places, and personal privacy information will be reflected and leaked out without the user's awareness, affecting the security of personal information. Mirror detection is a very challenging task due to the non-uniform size of mirrors and the presence of reflections. This paper proposes a frequency-domain based mirror detection method. Aiming at the reflection phenomenon existing on the mirror surface, we first proposed a frequency domain feature extraction module (FEM), which maps the multi-scale features of the mirror to the frequency domain, extracts the mirror features in the frequency domain, and suppresses the interference caused by the reflection of objects outside the mirror. In addition, for the edge inconsistency problem of the mirror surface, we propose a cross-level fusion module (CLFM) based on reverse attention, which fuses features of different levels and enhances image edge information. The experimental results show the good effect of our model.
- Wang, W., Xie, E., Li, X., Fan, D. P., Song, K., Liang, D., ... & Shao, L. 2022. Pvt v2: Improved baselines with pyramid vision transformer. Computational Visual Media, 8(3), 415-424. 2021.Google ScholarCross Ref
- Yang, X., Mei, H., Xu, K., Wei, X., Yin, B., & Lau, R. W. 2019. Where is my mirror ?. In Proceedings of the IEEE/CVF International Conference on Computer Vision(pp.8809-8818). 2019.Google ScholarCross Ref
- Lin, J., Wang, G., & Lau, R. W. 2020. Progressive mirror detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 3697-3705).Google ScholarCross Ref
- Guan, H., Lin, J., & Lau, R. W. 2022. Learning semantic associations for mirror detection. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 5941-5950).Google ScholarCross Ref
- He, R., Lin, J., & Lau, R. W. 2023, June. Efficient Mirror Detection via Multi-Level Heterogeneous Learning. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 37, No. 1, pp. 790-798).Google Scholar
- Lin, S., Zhang, Z., Huang, Z., Lu, Y., Lan, C., Chu, P., ... & Chen, Z. 2023. Deep frequency filtering for domain generalization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 11797-11807).Google ScholarCross Ref
- Sun, Y., Wang, S., Chen, C., & Xiang, T. Z. 2022. Boundary-guided camouflaged object detection. arXiv preprint arXiv:2207.00794.Google Scholar
- Chen, G., Shao, F., Chai, X., Chen, H., Jiang, Q., Meng, X., & Ho, Y. S. 2022. Modality-induced transfer-fusion network for RGB-D and RGB-T salient object detection. IEEE Transactions on Circuits and Systems for Video Technology, 33(4), 1787-1801.Google ScholarDigital Library
- Wei, J., Wang, S., & Huang, Q. 2020, April. F³Net: fusion, feedback and focus for salient object detection. In Proceedings of the AAAI conference on artificial intelligence (Vol. 34, No. 07, pp. 12321-12328).Google Scholar
- Wu, Z., Su, L., & Huang, Q. 2019. Cascaded partial decoder for fast and accurate salient object detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 3907-3916).Google ScholarCross Ref
- Chen, Z., Xu, Q., Cong, R., & Huang, Q. 2020, April. Global context-aware progressive aggregation network for salient object detection. In Proceedings of the AAAI conference on artificial intelligence (Vol. 34, No. 07, pp. 10599-10606).Google Scholar
- Liu, N., Zhang, N., Wan, K., Shao, L., & Han, J. 2021). Visual saliency transformer. In Proceedings of the IEEE/CVF international conference on computer vision (pp. 4722-4732).Google ScholarCross Ref
Index Terms
- Mirror Detection in Frequency Domain
Recommendations
GPU-based frequency domain volume rendering
SCCG '04: Proceedings of the 20th Spring Conference on Computer GraphicsFrequency domain volume rendering (FVR) is a volume rendering technique with lower computational complexity as compared to other techniques. In this paper the FVR algorithm is accelerated by factor of 17 by mapping the rendering stage to the GPU. The ...
Frequency domain normal map filtering
Filtering is critical for representing detail, such as color textures or normal maps, across a variety of scales. While MIP-mapping texture maps is commonplace, accurate normal map filtering remains a challenging problem because of nonlinearities in ...
A Theory Of Frequency Domain Invariants: Spherical Harmonic Identities for BRDF/Lighting Transfer and Image Consistency
This paper develops a theory of frequency domain invariants in computer vision. We derive novel identities using spherical harmonics, which are the angular frequency domain analog to common spatial domain invariants such as reflectance ratios. These ...
Comments