An image inpainting method for object removal based on difference degree constraint

Zhang, Lei; Chang, Minhui

doi:10.1007/s11042-020-09835-0

Cited by 6 publications

(3 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the matching rule is straightforward, it is likely to result in a mismatch error. The authors of [91] presented a difference degree constraint-based inpainting approach for object removal called Mean of Squared Differences. They utilized it to calculate the degree of separation between matching pixels in known places in the target patch and the exemplar patch.…”

Section: Object Removalmentioning

confidence: 99%

Pareto Optimized Large Mask Approach for Efficient and Background Humanoid Shape Removal

et al. 2023

View full text Add to dashboard Cite

We introduce deep learning-based methodology for removing unwanted human-like shapes in videos. The method uses Pareto-optimized Generative Adversarial Networks (GANs) technology, which is a novel contribution. The system automatically selects the Region of Interest (ROI) for each humanoid shape and uses a skeleton detection module to determine which humanoid shape to retain. The semantic masks of human like shapes are created using a semantic-aware occlusion-robust model that has four primary components: feature extraction, and local, global, and semantic branches. The global branch encodes occlusion-aware information to make the extracted features resistant to occlusion, while the local branch retrieves fine-grained local characteristics. A modified big mask inpainting approach is employed to eliminate a person from the image, leveraging Fast Fourier convolutions and utilizing polygonal chains and rectangles with unpredictable aspect ratios. The inpainter network takes the input image and the mask to create an output image excluding the background humanoid shapes. The generator uses an encoder-decoder structure with included skip connections to recover spatial information and dilated convolution and squeeze and excitation blocks to make the regions behind the humanoid shapes consistent with their surroundings. The discriminator avoids dissimilar structure at the patch scale, and the refiner network catches features around the boundaries of each background humanoid shape. The method is evaluated on two video object segmentation datasets (DAVIS and YouTube-VOS) and a database of 66 distinct video sequences of people behind a desk in an office environment. The efficiency was assessed using the Structural Similarity Index Measure (SSIM), Frechet Inception Distance (FID), and Learned Perceptual Image Patch Similarity (LPIPS) metrics and showed promising results in fully automated background person removal task.

show abstract

Section: Object Removalmentioning

confidence: 99%

Pareto Optimized Large Mask Approach for Efficient and Background Humanoid Shape Removal

et al. 2023

View full text Add to dashboard Cite

show abstract

“…After selecting the main thing, the task of the grab cut algorithm is to eliminate objects that are not selected by marking them as background. When objects are not needed, the object removal process may be performed [18], [19]. If a part of the main object is missing, you can return the thing by placing a marker on the missing piece.…”

Section: Features Selectionmentioning

confidence: 99%

Comparison Study of Convolutional Neural Network Architecture in Aglaonema Classification

Mulyani

Septiangraini

Muhammad

et al. 2022

IJECS

View full text Add to dashboard Cite

Convolutional Neural Network (CNN) is very good at classifying images. To measure the best CNN architecture, a study must be done against real-case scenarios. Aglaonema, one of the plants with high similarity, is chosen as a test case to compare CNN architecture. In this study, a classification process was carried out on five classes of Aglaonema imagery by comparing five architectures from the Convolutional Neural Network (CNN) method: LeNet, AlexNet, VGG16, Inception V3, and ResNet50. The total dataset used is 500 image data, with the distribution of training data by 80% and test data by 20%. The segmentation process is performed using the Grabcut algorithm by separating the foreground and background. To build a model for CNN architecture using Google Colab and Google Drive storage. The results of the tests carried out on five classes of Aglaonema images obtained the best accuracy, precision, and recall results on the Inception V3 architecture with values of 92.8%, 93%, and 92.8%. The CNN architecture has the highest level of accuracy in classifying aglaonema plant types based on images. This study seeks to close research gaps, contribute to the field of research, and serve as a platform for primary prevention research.

show abstract

“…The authors Janardhana Rao et al [13][14][15] introduced an improved approach for priority computation, which involves the integration of regularization factors and adaptive coefficients. Zhang et al [16] employed the combination of mean squared diffrerence and square of mean differences as a similarity metric to find the exemplar patch. The video inpainting using exemplar based methods are also evevated in recent years [17][18][19][20][21][22][23][24].…”

Section: Introductionmentioning

confidence: 99%

Development of Image Inpainting for object removal using Enhanced Patch Priority and Matching Measures

Rao,

Revathi,

Bhanusree

et al. 2024

E3S Web Conf.

View full text Add to dashboard Cite

Image inpainting can be used to fix broken images and get rid of distracting elements. In exemplar based methods, patch priority computation and exemplar patch selection are crucial to the success of image inpainting technique. The dropping effect occurred in the highest patch priority computation and matching error in the best patch selection are the major issues in the exemplar inpaint methods. In this paper, the enhanced priority calculation technique is employed to avoid the dropping effect and introduced the new similarity measuring process, Mean Squared Error (MSD). The efficacy of the proposed techniques is estimated by comparing with the available methods in the literature qualitatively. It shows that proposed methods outperformed existing techniques.

show abstract

An image inpainting method for object removal based on difference degree constraint

Cited by 6 publications

References 37 publications

Pareto Optimized Large Mask Approach for Efficient and Background Humanoid Shape Removal

Pareto Optimized Large Mask Approach for Efficient and Background Humanoid Shape Removal

Comparison Study of Convolutional Neural Network Architecture in Aglaonema Classification

Development of Image Inpainting for object removal using Enhanced Patch Priority and Matching Measures

Contact Info

Product

Resources

About