VeIGAN: Vectorial Inpainting Generative Adversarial Network for Depth Maps Object Removal

Matias, Lucas P. N.; Sons, M. Witmark; Souza, Jefferson R.; Wolf, Denis F.; Stiller, Christoph

doi:10.1109/ivs.2019.8814157

Cited by 3 publications

(23 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several approaches for depth image inpainting exist, many of which focus on utilizing available corresponding RGB data as context for the inference of the missing depth information [42][43][44][45]. Other works approach the problem by training models that attempt to minimize the difference between the surface normals of the completed depth image and its ground truth [45,46]. inpainting methods, but lacks experiments to evaluate the proposed architecture.…”

Section: Background and Contextmentioning

confidence: 99%

“…Inspired by several works that employ surface normals to depth image inpainting and generation [45,46,[59][60][61], we proposed the application of this concept to joint RGB-D image inpainting.…”

Section: Research Objective 12 Define An Architecture That Is Capable Of Handling the Multimodal Characteristics Of Rgb-d Imagesmentioning

confidence: 99%

“…In particular, we proposed the usage of a surface normal loss [46] function, which we demonstrated to improve the reproduction of the desired properties of the depth channel such as smoothness. Moreover, we replaced the contextual attention module of the base framework with the contextual surface attention module [46] and showed how this module benefits from auxiliary surface normal information. Lastly, we proposed the addition of surface normal information to the input of the discriminator network, which we determined to cause a deterioration of the visual quality of the inpainted results.…”

Section: Research Objective 12 Define An Architecture That Is Capable Of Handling the Multimodal Characteristics Of Rgb-d Imagesmentioning

confidence: 99%

“…Image inpainting methods focused on denoising can be applied for the removal of artifacts of this type [43,103]. Moreover, depth image inpainting techniques can be applied to object removal and surface completion in depth images [46,94]. In this way, applications that use depth data can greatly profit from depth image inpainting.…”

Section: Depth Image Inpaintingmentioning

confidence: 99%

“…Existing RGB image inpainting methods have previously been applied to depth image inpainting [46,[118][119][120] with reasonable success. However, in practice, these methods often fail to address the statistical properties of depth image inpainting involving depth continuity, surface relief, and local feature preservation [121].…”

Section: Depth Image Inpaintingmentioning

confidence: 99%

See 4 more Smart Citations

Generative RGB-D Face Completion for Head-Mounted Display Removal

Numan

Haar²,

César

2021

2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)

View full text Add to dashboard Cite

My supervisors, Frank ter Haar and Pablo Cesar, for providing invaluable guidance through their constructive feedback, great patience and extensive knowledge.• My colleagues at the Intelligent Imaging department of TNO, for their exciting suggestions, stimulating feedback and infectious enthusiasm.• My mom and sister, for their endless love, support and encouragement.• My grandparents, for always believing in me.

show abstract

Section: Background and Contextmentioning

confidence: 99%

Section: Research Objective 12 Define An Architecture That Is Capable Of Handling the Multimodal Characteristics Of Rgb-d Imagesmentioning

confidence: 99%

Section: Research Objective 12 Define An Architecture That Is Capable Of Handling the Multimodal Characteristics Of Rgb-d Imagesmentioning

confidence: 99%

Section: Depth Image Inpaintingmentioning

confidence: 99%

Section: Depth Image Inpaintingmentioning

confidence: 99%

See 3 more Smart Citations

Generative RGB-D Face Completion for Head-Mounted Display Removal

Numan

Haar²,

César

2021

2021 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW)

View full text Add to dashboard Cite

show abstract

Binocular Stereovision

Zhang

2024

3D Computer Vision

View full text Add to dashboard Cite

Large-Scale 3D Reconstruction from Multi-View Imagery: A Comprehensive Review

Luo,

Zhang,

Liu

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Three-dimensional reconstruction is a key technology employed to represent virtual reality in the real world, which is valuable in computer vision. Large-scale 3D models have broad application prospects in the fields of smart cities, navigation, virtual tourism, disaster warning, and search-and-rescue missions. Unfortunately, most image-based studies currently prioritize the speed and accuracy of 3D reconstruction in indoor scenes. While there are some studies that address large-scale scenes, there has been a lack of systematic comprehensive efforts to bring together the advancements made in the field of 3D reconstruction in large-scale scenes. Hence, this paper presents a comprehensive overview of a 3D reconstruction technique that utilizes multi-view imagery from large-scale scenes. In this article, a comprehensive summary and analysis of vision-based 3D reconstruction technology for large-scale scenes are presented. The 3D reconstruction algorithms are extensively categorized into traditional and learning-based methods. Furthermore, these methods can be categorized based on whether the sensor actively illuminates objects with light sources, resulting in two categories: active and passive methods. Two active methods, namely, structured light and laser scanning, are briefly introduced. The focus then shifts to structure from motion (SfM), stereo matching, and multi-view stereo (MVS), encompassing both traditional and learning-based approaches. Additionally, a novel approach of neural-radiance-field-based 3D reconstruction is introduced. The workflow and improvements in large-scale scenes are elaborated upon. Subsequently, some well-known datasets and evaluation metrics for various 3D reconstruction tasks are introduced. Lastly, a summary of the challenges encountered in the application of 3D reconstruction technology in large-scale outdoor scenes is provided, along with predictions for future trends in development.

show abstract

VeIGAN: Vectorial Inpainting Generative Adversarial Network for Depth Maps Object Removal

Cited by 3 publications

References 23 publications

Generative RGB-D Face Completion for Head-Mounted Display Removal

Generative RGB-D Face Completion for Head-Mounted Display Removal

Binocular Stereovision

Large-Scale 3D Reconstruction from Multi-View Imagery: A Comprehensive Review

Contact Info

Product

Resources

About