Fast multi-view image rendering method based on reverse search for matching

Lin, Jianpu; Wang, Weixing; Yao, Jianmin; Guo, Tao; Chen, Enguo; Yan, Qun

doi:10.1016/j.ijleo.2018.12.003

Cited by 5 publications

(3 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Lai et al [29] use a depth-reliability-map-based occlusion aware approach to create a segmentation mask, that indicates where the information in the novel synthesized view should be blended. Lin et al [30] propose a fast multi-view image rendering method that uses a pixel mapping information to derive a rendering image. This method reduces rendering time and memory size comparing to a conventional DIBR method, however it has the same effect in terms of synthesized image quality as DIBR.…”

Section: Other Algorithmic Methodsmentioning

confidence: 99%

Hybrid dual stream blender for wide baseline view synthesis

Hobloss

Zhang

Lathuilière

et al. 2021

Signal Processing: Image Communication

View full text Add to dashboard Cite

Section: Other Algorithmic Methodsmentioning

confidence: 99%

Hybrid dual stream blender for wide baseline view synthesis

Hobloss

Zhang

Lathuilière

et al. 2021

Signal Processing: Image Communication

View full text Add to dashboard Cite

“…Subsequently, surface reconstruction techniques such as Poisson reconstruction are often applied as post-processing steps to generate the final 3D model. The quality of reconstruction heavily relies on corresponding matching [16] and is not suitable for objects lacking rich textures.…”

Section: Multi-view 3d Reconstruction Traditional Multi-view 3d Recon...mentioning

confidence: 99%

P‐4.15: Multi‐Viewpoint Glasses‐Free 3D Display Technology Based on Representation Fusion NeRF

Huang,

Surman,

Sun

2024

Symp Digest of Tech Papers

View full text Add to dashboard Cite

The intense visual impact brought about by glasses‐free 3D display has changed people's perception of traditional display, garnering increasing attention and research, and it is anticipated to reach a market size of billions of dollars within the next five years. However, a challenge lies in the scarcity of display content resources and their high degree of manual customization, leading to elevated costs for three‐dimensional (3D) display deployment, hindering widespread adoption. We propose a novel approach that combines a three‐dimensional reconstruction technique, specifically Neural Radiance Fields (NeRF), with multi‐viewpoint synthesis. Meanwhile, we integrate the Signed Distance Function (SDF) into the volume rendering framework of the reconstruction algorithm for representation fusion optimization. Finally, we utilize lenticular lens arrays to realize the 3D display effect. Experimental results show that our method achieves superior high‐quality reconstruction compared to previous algorithms and provides efficient display of custom‐captured scenes.

show abstract

“…Recently, deep learning-based approaches have achieved a remarkable traction, which also motivate the development of new deep learning-based approaches [18][19][20][21] for low-level image processing tasks, including super-resolution [22,23], rain removal [24,25], hyperspectral image [26,27], and so on. Shen et al [28] proposed the MSR-net to enhance the low-light image by learning a mapping from the low-light images to normal-light images.…”

Section: Introductionmentioning

confidence: 99%

Single Low-Light Ghost-Free Image Enhancement via Deep Retinex Model

Liu¹,

Lv²,

Wang³

et al. 2021

KSII TIIS

View full text Add to dashboard Cite

Low-light image enhancement is a key technique to overcome the quality degradation of photos taken under scotopic vision illumination conditions. The degradation includes low brightness, low contrast, and outstanding noise, which would seriously affect the vision of the human eye recognition ability and subsequent image processing. In this paper, we propose an approach based on deep learning and Retinex theory to enhance the low-light image, which includes image decomposition, illumination prediction, image reconstruction, and image optimization. The first three parts can reconstruct the enhanced image that suffers from lowresolution. To reduce the noise of the enhanced image and improve the image quality, a superresolution algorithm based on the Laplacian pyramid network is introduced to optimize the image. The Laplacian pyramid network can improve the resolution of the enhanced image through multiple feature extraction and deconvolution operations. Furthermore, a combination loss function is explored in the network training stage to improve the efficiency of the algorithm. Extensive experiments and comprehensive evaluations demonstrate the strength of the proposed method, the result is closer to the real-world scene in lightness, color, and details. Besides, experiments also demonstrate that the proposed method with the single low-light image can achieve the same effect as multi-exposure image fusion algorithm and no ghost is introduced.

show abstract

Fast multi-view image rendering method based on reverse search for matching

Cited by 5 publications

References 10 publications

Hybrid dual stream blender for wide baseline view synthesis

Hybrid dual stream blender for wide baseline view synthesis

P‐4.15: Multi‐Viewpoint Glasses‐Free 3D Display Technology Based on Representation Fusion NeRF

Single Low-Light Ghost-Free Image Enhancement via Deep Retinex Model

Contact Info

Product

Resources

About