Lightweight high-resolution Subject Matting in the Real World (2312.07100v1)
Abstract: Existing saliency object detection (SOD) methods struggle to satisfy fast inference and accurate results simultaneously in high resolution scenes. They are limited by the quality of public datasets and efficient network modules for high-resolution images. To alleviate these issues, we propose to construct a saliency object matting dataset HRSOM and a lightweight network PSUNet. Considering efficient inference of mobile depolyment framework, we design a symmetric pixel shuffle module and a lightweight module TRSU. Compared to 13 SOD methods, the proposed PSUNet has the best objective performance on the high-resolution benchmark dataset. Evaluation results of objective assessment are superior compared to U$2$Net that has 10 times of parameter amount of our network. On Snapdragon 8 Gen 2 Mobile Platform, inference a single 640$\times$640 image only takes 113ms. And on the subjective assessment, evaluation results are better than the industry benchmark IOS16 (Lift subject from background).
- “Saliency detection via graph-based manifold ranking,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2013, pp. 3166–3173.
- “Saliency optimization from robust background detection,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2014, pp. 2814–2821.
- “Salient object detection via objectness measure,” in 2015 IEEE international conference on image processing (ICIP). IEEE, 2015, pp. 4481–4485.
- “Minimum barrier salient object detection at 80 fps,” in Proceedings of the IEEE international conference on computer vision, 2015, pp. 1404–1412.
- “Real-esrgan: Training real-world blind super-resolution with pure synthetic data,” in Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshops, October 2021, pp. 1905–1914.
- “Rgb-d saliency detection via cascaded mutual information minimization,” in International Conference on Computer Vision (ICCV), 2021.
- “A de-raining semantic segmentation network for real-time foreground segmentation,” Journal of Real-Time Image Processing, vol. 18, pp. 873–887, 2021.
- “U2-net: Going deeper with nested u-structure for salient object detection,” 2020, vol. 106, p. 107404.
- “Real-time single image and video super-resolution using an efficient sub-pixel convolutional neural network,” in Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 1874–1883.
- “Mobilesal: Extremely efficient rgb-d salient object detection,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 44, no. 12, pp. 10261–10269, 2022.
- “Tracer: Extreme attention guided salient object tracing network,” arXiv preprint arXiv:2112.07380, 2021.
- “A simple pooling-based design for real-time salient object detection,” in IEEE CVPR, 2019.
- “P2t: Pyramid pooling transformer for scene understanding,” IEEE Transactions on Pattern Analysis and Machine Intelligence, pp. 1–12, 2022.
- “Highly accurate dichotomous image segmentation,” in ECCV, 2022.
- “Boundary-aware segmentation network for mobile and web applications,” arXiv preprint arXiv:2101.04704, 2021.
- “Revisiting image pyramid structure for high resolution salient object detection,” 2022.
- “U-net: Convolutional networks for biomedical image segmentation,” 2015.
- “Towards high-resolution salient object detection,” 2019.
- “A benchmark dataset and evaluation methodology for video object segmentation,” in 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2016, pp. 724–732.
- “Pyramid grafting network for one-stage high resolution saliency detection,” in CVPR, 2022.
- “Image quality assessment: from error visibility to structural similarity,” IEEE transactions on image processing, vol. 13, no. 4, pp. 600–612, 2004.
- “Adam: A method for stochastic optimization,” 2017.
- “Learning to detect salient objects with image-level supervision,” in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2017, pp. 3796–3805.
- Peng Liu (372 papers)
- Fanyi Wang (18 papers)
- Jingwen Su (7 papers)
- Yanhao Zhang (33 papers)
- Guojun Qi (15 papers)