Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
97 tokens/sec
GPT-4o
53 tokens/sec
Gemini 2.5 Pro Pro
44 tokens/sec
o3 Pro
5 tokens/sec
GPT-4.1 Pro
47 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Diverse Image Harmonization (2407.15481v1)

Published 22 Jul 2024 in cs.CV

Abstract: Image harmonization aims to adjust the foreground illumination in a composite image to make it harmonious. The existing harmonization methods can only produce one deterministic result for a composite image, ignoring that a composite image could have multiple plausible harmonization results due to multiple plausible reflectances. In this work, we first propose a reflectance-guided harmonization network, which can achieve better performance with the guidance of ground-truth foreground reflectance. Then, we also design a diverse reflectance generation network to predict multiple plausible foreground reflectances, leading to multiple plausible harmonization results. The extensive experiments on the benchmark datasets demonstrate the effectiveness of our method.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (50)
  1. Pixelnn: Example-based image synthesis. arXiv preprint arXiv:1708.05349, 2017.
  2. Deep image-based illumination harmonization. In CVPR, 2022.
  3. Piigan: generative adversarial networks for pluralistic image inpainting. IEEE Access, 8, 2020.
  4. A structure-preserving and illumination-consistent cycle framework for image harmonization. IEEE Transactions on Multimedia, 2023.
  5. Dense pixel-to-pixel harmonization via continuous image representation. arXiv preprint arXiv:2303.01681, 2023.
  6. Infogan: Interpretable representation learning by information maximizing generative adversarial nets. NIPS, 2016.
  7. Dovenet: Deep image harmonization via domain verification. In CVPR, 2020.
  8. High-resolution image harmonization via collaborative dual transformations. In CVPR, 2022.
  9. Improving the harmony of the composite image by spatial-separated attention module. IEEE Transactions on Image Processing, 29:4759–4771, 2020.
  10. Adversarial feature learning. arXiv preprint arXiv:1605.09782, 2016.
  11. Adversarially learned inference. arXiv preprint arXiv:1606.00704, 2016.
  12. Pct-net: Full resolution image harmonization using pixel-wise color transformations. In CVPR, 2023.
  13. Image harmonization with transformer. In ICCV, 2021a.
  14. Intrinsic image harmonization. In CVPR, 2021b.
  15. Transformer for image harmonization and beyond. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2022.
  16. Scs-co: Self-consistent style contrastive learning for image harmonization. In CVPR, 2022.
  17. Image harmonization with attention-based deep feature modulation. In BMVC, 2020.
  18. Reducing the dimensionality of data with neural networks. science, 313(5786):504–507, 2006.
  19. Ssh: A self-supervised framework for image harmonization, 2021.
  20. Analyzing and improving the image quality of stylegan. In CVPR, 2020.
  21. Harmonizer: Learning to perform white-box image and video harmonization. In ECCV, 2022.
  22. A variational framework for retinex. International Journal of computer vision, 52(1):7–23, 2003.
  23. Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
  24. Using color compatibility for assessing image realism. In ICCV, 2007.
  25. Autoencoding beyond pixels using a learned similarity metric. In ICML, 2016.
  26. Spatial-separated curve rendering network for efficient and high-resolution image harmonization. In ECCV, 2022.
  27. Region-aware adaptive instance normalization for image harmonization. In CVPR, 2021.
  28. Image inpainting for irregular holes using partial convolutions. In ECCV, 2018.
  29. Reduce information loss in transformers for pluralistic image inpainting. In CVPR, 2022.
  30. Lemart: Label-efficient masked region transform for image harmonization. In CVPR, 2023.
  31. Least squares generative adversarial networks. In ICCV, 2017.
  32. Conditional generative adversarial nets. arXiv preprint arXiv:1411.1784, 2014.
  33. Making images real again: A comprehensive survey on deep image composition. arXiv preprint arXiv:2106.14490, 2021.
  34. Color transfer between images. IEEE Computer Graphics and Applications, 21(5):34–41, 2001.
  35. Semantic-guided multi-mask image harmonization. In ECCV, 2022.
  36. Learning global-aware kernel for image harmonization. arXiv preprint arXiv:2305.11676, 2023.
  37. Foreground-aware semantic representations for image harmonization, 2020.
  38. Multi-scale image harmonization. ACM Transactions on Graphics, 29(4):125:1–125:10, 2010.
  39. Deep image harmonization. In CVPR, 2017.
  40. Interactive portrait harmonization. arXiv preprint arXiv:2203.08216, 2022.
  41. Neural discrete representation learning. NIPS, 2017.
  42. High-fidelity pluralistic image completion with transformers. In ICCV, 2021.
  43. Semi-supervised parametric real-world image harmonization. In CVPR, 2023.
  44. Composite photograph harmonization with complete background cues. In ACM MM, 2022.
  45. Dccf: Deep comprehensible color filter learning framework for high-resolution image harmonization. In ECCV, 2022.
  46. Understanding and improving the realism of image composites. ACM Transactions on Graphics, 31(4):84:1–84:10, 2012.
  47. Self-supervised outdoor scene relighting. In ECCV, 2020.
  48. Pluralistic image completion. In CVPR, 2019.
  49. Learning a discriminative model for the perception of realism in composite images. In ICCV, 2015.
  50. Toward multimodal image-to-image translation. NIPS, 2017.
User Edit Pencil Streamline Icon: https://streamlinehq.com
Authors (4)
  1. Xinhao Tao (5 papers)
  2. Tianyuan Qiu (1 paper)
  3. Junyan Cao (35 papers)
  4. Li Niu (79 papers)

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now