MiniGAN: Toward Informative and Uninformative Image Transferring
Source abstract
Abstract
This article proposes a generative adversarial networks (MiniGAN) to tackle both informative and uninformative image transferring. The generator of MiniGAN is based on the structure of StyleGANv2, in which the encoder and style transform block are proposed to extract the high-level feature maps of the source image and capture the latent representation of the target image, respectively. This information guides the generator for the final image generation. The proposed MiniGAN outperforms other models in style transferring while preserving the color information on the informative images. To test the performance of MiniGAN on the uninformative images, a new data set consisting of 10,000 fashion hand drawings is proposed. Extensive experiments and detailed analysis are presented to demonstrate the performance of MiniGAN.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Generative Adversarial Networks and Image Synthesis · Digital Media Forensic Detection · Advanced Image Processing Techniques
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
The Hong Kong Polytechnic University, Hung Hom, Hong Kong
The Chinese University of Hong Kong, Shatin, Hong Kong
Funding
Project Code: RP 3-1
References (41)
Show source-provided references
1. Isola P · 1125
2. Zhu J-Y · 2223
3. Cao Y · 151
4. Nazeri K · 85
5. Johnson J · 694
6. Sanakoyeu A · 698
7. Park T · 319
8. Karras T · 8110
9. Goodfellow IJ, Pouget-Abadie J, Mirza M, et al. Generative adversarial networks. arXiv preprintarXiv:1406.2661, 2014, https://arxiv.org/abs/1406.2661
10. Radford A, Metz L, Chintala S. Unsupervised representation learning with deep convolutional generative adversarial networks. arXiv preprintarXiv:1511.06434, 2015, https://arxiv.org/abs/1511.06434
11. Brock A, Donahue J, Simonyan K. Large scale GAN training for high fidelity natural image synthesis. arXiv preprintarXiv:1809.11096, 2018, https://arxiv.org/abs/1809.11096
12. Karras T, Aila T, Laine S, et al. Progressive growing of GANs for improved quality, stability, and variation. arXiv preprintarXiv:1710.10196, 2017, https://arxiv.org/abs/1710.10196
13. Karras T · 4401
14. Arjovsky M · 2017 · 70 · 214
15. Gulrajani I, Ahmed F, Arjovsky M, et al. Improved training of Wasserstein GANs. arXiv preprintarXiv:1704.00028, 2017, https://arxiv.org/abs/1704.00028
16. Miyato T, Kataoka T, Koyama M, et al. Spectral normalization for generative adversarial networks. arXiv preprintarXiv:1802.05957, 2018, https://arxiv.org/abs/1802.05957
17. Gatys LA · 2414
18. Gatys LA · 3985
19. Huang X · 1501
20. Li Y, Fang C, Yang J, et al. Universal style transfer via feature transforms. arXiv preprintarXiv:1705.08086, 2017, https://arxiv.org/abs/1705.08086
21. Li Y · 453
22. Sheng L · 8242
23. Yoo J · 9036
24. Jing Y · 238
25. Yao Y · 1467
26. Liu M-Y, Breuel T, Kautz J. Unsupervised image-to-image translation networks. arXiv preprintarXiv:1703.00848, 2017, https://arxiv.org/abs/1703.00848
27. Huang X · 172
28. 10.1007/s11263-019-01284-z10.1007/s11263-019-01284-z ↗
29. Richardson E, Alaluf Y, Patashnik O, et al. Encoding in style: A StyleGAN encoder for image-to-image translation. arXiv preprintarXiv:2008.00951, 2020, https://arxiv.org/abs/2008.00951
30. He K · 770
31. Svoboda J · 13816
32. Ronneberger O · 234
33. Mao X · 2794
34. Wang X · 8090
35. Lu C · 65
36. Cho W · 10639
37. Kingma DP, Ba J. Adam: A method for stochastic optimization. arXiv preprintarXiv:1412.6980, 2014, https://arxiv.org/abs/1412.6980
38. 10.1109/TPAMI.2017.272300910.1109/TPAMI.2017.2723009 ↗
39. Heusel M, Ramsauer H, Unterthiner T, et al. GANs trained by a two time-scale update rule converge to a local Nash equilibrium. arXiv preprintarXiv:1706.08500, 2017, https://arxiv.org/abs/1706.08500
40. Szegedy C · 2818
41. Zhang R · 586