Structure guided texture inpainting through multi-scale patches and global optimization for image completion

Chen, XiaoWu; Zhou, Bin; Guo, Yueping; Xu, Fang; Zhao, Qi

doi:10.1007/s11432-012-4772-7

Cited by 7 publications

(5 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this work, we use graphs and histograms of their subgraphs for discovering family photos. Interestingly, there have been many graphbased approaches for image extrapolation [19], interpolation [20], image segmentation [21], representations [22], etc. While these applications are not directly related to our classification problem, utilizing histograms of subgraphs could be useful in these applications, e.g., better graph matching for extrapolation.…”

Section: Graph-based Image Editingmentioning

confidence: 99%

Discriminative subgraphs for discovering family photos

2016

View full text Add to dashboard Cite

We propose to use discriminative subgraphs to discover family photos from group photos in an efficient and effective way. Group photos are represented as face graphs by identifying social contexts such as age, gender, and face position. The previous work utilized bag-of-word models and considered frequent subgraphs from all group photos as features for classification. This approach, however, produces numerous subgraphs, resulting in high dimensions. Furthermore, some of them are not discriminative. To solve these issues, we adopt a state-of-the-art, frequent subgraph mining method that removes nondiscriminative subgraphs.We also use TF-IDF normalization, which is more suitable for the bag-ofword model. To validate our method, we experiment in two datasets.Our method shows consistently better performance, higher accuracy in lower feature dimensions, compared to the previous method. We also integrate our method with the recent Microsoft face recognition API and release it in a public website.

show abstract

Section: Graph-based Image Editingmentioning

confidence: 99%

Discriminative subgraphs for discovering family photos

2016

View full text Add to dashboard Cite

show abstract

“…The sampling process includes two categories: pixel-based algorithms [12][13][14] and patch-based algorithms [15,16]. Similarly, there are some approaches in image completion working on the pixel basis [17][18][19] while the fragmentbased methods [1][2][3][4][5] could obtain better quality and performance.…”

Section: Related Workmentioning

confidence: 99%

“…At present, the most popular techniques are the fragment/exemplar based image completion methods [1][2][3][4][5] and a range of improvements. These approaches vary essentially in three aspects.…”

Section: Introductionmentioning

confidence: 99%

单幅图像上透视场景的图像填补

Hao

Chen

et al. 2015

Sci. China Inf. Sci.

View full text Add to dashboard Cite

Although image completion has been used in many image filling and editing problems for a long time, it is rarely applied in the scenario regarding the perspective effect which is demonstrated largely in the design of urban architectures. Thus this paper proposes a method to capture the perspective information automatically based on a single image and employs it to image completion tasks effectively. Actually it is difficult to automatize the extraction from only a single image, but it has been observed that architectures usually contain plenty of repetitions and symmetries which could be described through periodicity. Therefore, our approach first determines the initial vanishing points through the cascaded hough transformation. Then, a refinement scheme based on SIFT feature matching is proposed to obtain the accurate vanishing points. At last, a good perspective mesh is generated from the geometry calculation. The experiments show that this method could be used to automatically extract perspective information from a single image and recover the image in a visually plausible way.

show abstract

“…In the Structure Propagation system [7][8][9][10][11][12], the system automatically finds the salient structural information around the unknown regions, and extends them into unknown regions to construct assistant curves. After synthesizing the missing structural information along the assistant curves, the remaining unknown regions can be filled using patch-based texture synthesis and the original images will be restored.…”

Section: B Image Completingmentioning

confidence: 99%

Perfect Group Photo Synthesis

Qiu

et al. 2015

2015 International Conference on Virtual Reality and Visualization (ICVRV)

View full text Add to dashboard Cite

In a group photo, individual facial expressions can fatally affect the quality of the whole photo. However, it is a laborious task for most people to synthesize a perfect group photo using current photo editing software. To solve this problem, an automatic perfect group photo synthesis algorithm is proposed in this paper by judging facial expressions and replacing faces. This algorithm proceeds in four steps. First, faces in group photos are recognized and saved. Then, by evaluating facial expressions for each person, one face with the best expression is selected. Finally, a final synthesizing group photo is produced by color transferring and image completing. Experimental results prove that this algorithm is effective and can be used to synthesize better group photos automatically.

show abstract

Structure guided texture inpainting through multi-scale patches and global optimization for image completion

Cited by 7 publications

References 26 publications

Discriminative subgraphs for discovering family photos

Discriminative subgraphs for discovering family photos

单幅图像上透视场景的图像填补

Perfect Group Photo Synthesis

Contact Info

Product

Resources

About