Semantic-Aware Face Deblurring with Pixel-Wise Projection Discriminator
- 주제(키워드) 영상 복원 , 영상 생성 모델 , 디블러링
- 주제(DDC) 006.31
- 발행기관 아주대학교
- 지도교수 허용석
- 발행년도 2023
- 학위수여년월 2023. 2
- 학위명 석사
- 학과 및 전공 일반대학원 인공지능학과
- 실제URI http://www.dcollection.net/handler/ajou/000000032548
- 본문언어 영어
- 저작권 아주대학교 논문은 저작권에 의해 보호받습니다.
초록/요약
Most recent face deblurring methods have leveraged the distribution modeling ability of generative adversarial networks (GANs) to impose a constraint that the deblurred image should follow the distribution of sharp ground-truth images. However, generating sharp face images with high fidelity and realistic properties from a blurry face image remains challenging under the GAN framework. To this end, we focus on modeling the joint distribution of sharp face images and segmentation label maps for face image deblurring in a GAN framework. We propose a semantic-aware pixel-wise projection (SAPP) discriminator that models pixel-label matching with semantic label map information and generates a pixel-wise probability map of realness for the input image as well as a per-image probability. Moreover, we introduce a prediction-weighted (PW) loss to focus on erroneous pixels in the output of the decoder, using per-pixel real/fake probability map to re-weight the contribution of each pixel in the decoder. Furthermore, we present a coarse-to-fine training technique for the generator, which encourages the generator to focus on global consistency in the early training stages and local details in the later stages. Extensive experimental results show that our method outperforms existing methods both quantitatively and qualitatively in terms of perceptual image quality.
more목차
1. Introduction 1
2.Related Works 6
2.1 Single Image Deblurring 6
2.1.1 General Deblurring 6
2.1.2 Face Deblurring 7
2.2 Generative Adversarial Networks 8
2.2.1 Conditional GANs 9
2.2.2 U-Net based GANs 9
3. Proposed Method 11
3.1 Semantic-Aware Pixel-Wise Projection Discriminator 12
3.2 Discriminator training 14
3.3 Generator training 16
4. Experiments and Results 19
4.1 Experimental Details 19
4.1.1 Datasets 19
4.1.2 Implementation Details 20
4.1.3 Evaluation Metrics 20
4.2 Comparisons on MSPL dataset 21
4.3 Comparisons on real blurred images 25
4.4 Execution time and Face Verification 26
4.5 Ablation Study 27
5. Conclusions 30
Reference 31
