Joint Texture and Geometry Optimization for RGB-D Reconstruction

Fu, Yanping; Yan, Qingan; Liao, Jie; Xiao, Chunxia

doi:10.1109/cvpr42600.2020.00599

Cited by 33 publications

(24 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the future, we are going to combine techniques as detail‐preserved surface reconstruction, 29 texture mapping, 30,31 illumination decomposition, 32 bidirectional reflectance distribution function reconstruction, 33 and MVS for specular highlights 34 to reconstruct more realistic 3D models.…”

Section: Resultsmentioning

confidence: 99%

Dense multiview stereo based on image texture enhancement

Liao

Wei

et al. 2020

Computer Animation & Virtual

Self Cite

View full text Add to dashboard Cite

In this paper, we propose a novel Multiview Stereo (MVS) method which can effectively estimate geometry in low‐textured regions. Conventional MVS algorithms predict geometry by performing dense correspondence estimation across multiple views under the constraint of epipolar geometry. As low‐textured regions contain less feature information for reliable matching, estimating geometry for low‐textured regions remains hard work for previous MVS methods. To address this issue, we propose an MVS method based on texture enhancement. By enhancing texture information for each input image via our multiscale bilateral decomposition and reconstruction algorithm, our method can estimate reliable geometry for low‐textured regions that are intractable for previous MVS methods. To densify the final output point cloud, we further propose a novel selective joint bilateral propagation filter, which can effectively propagate reliable geometry estimation to neighboring unpredicted regions. We validate the effectiveness of our method on the ETH3D benchmark. Quantitative and qualitative comparisons demonstrate that our method can significantly improve the quality of reconstruction in low‐textured regions.

show abstract

Section: Resultsmentioning

confidence: 99%

Dense multiview stereo based on image texture enhancement

Liao

Wei

et al. 2020

Computer Animation & Virtual

Self Cite

View full text Add to dashboard Cite

show abstract

“…Second, our framework separates the 3D registration part and texture map generation part. We expect better results if the geometry and texture are optimized jointly [FYLX20].…”

Section: Discussionmentioning

confidence: 95%

Spatiotemporal Texture Reconstruction for Dynamic Objects Using a Single RGB‐D Camera

Kim

Nam

et al. 2021

Computer Graphics Forum

View full text Add to dashboard Cite

Our approach reconstructs a time-varying (spatiotemporal) texture map for a dynamic object using partial observations obtained by a single RGB-D camera. The frontal and rear views (top and bottom rows) of the geometry at two frames are shown in the middle left. Compared to the global texture atlas-based approach [KKPL19], our method produces more appealing appearance changes of the object. Please see the supplementary video for better visualization of time-varying textures.

show abstract

“…Due to errors in the camera alignment or in the surface geometry, blurry textures or patch seams affect results and additional texture alignment procedures have been proposed [42,6,86,20,27,90,41,83] to tackle these problems. Fu et al [24] jointly refine geometry, texture, and camera poses after classical Kinect-Fusion [54] reconstruction. Better texture mapping results have been achieved with an optical flow-like correction in texture space [20,90,25], patch-based optimization [7], or via 2D perspective warp techniques [39].…”

Section: Appearance Estimationmentioning

confidence: 99%

DeepSurfels: Learning Online Appearance Fusion

Mihajlovic¹,

Weder²,

Pollefeys³

et al. 2020

Preprint

View full text Add to dashboard Cite

We present DeepSurfels, a novel hybrid scene representation for geometry and appearance information. DeepSurfels combines explicit and neural building blocks to jointly encode geometry and appearance information. In contrast to established representations, DeepSurfels better represents high-frequency textures, is well-suited for online updates of appearance information, and can be easily combined with machine learning methods. We further present an end-to-end trainable online appearance fusion pipeline that fuses information provided by RGB images into the proposed scene representation and is trained using selfsupervision imposed by the reprojection error with respect to the input images. Our method compares favorably to classical texture mapping approaches as well as recently proposed learning-based techniques. Moreover, we demonstrate lower runtime, improved generalization capabilities, and better scalability to larger scenes compared to existing methods.

show abstract

Joint Texture and Geometry Optimization for RGB-D Reconstruction

Cited by 33 publications

References 35 publications

Dense multiview stereo based on image texture enhancement

Dense multiview stereo based on image texture enhancement

Spatiotemporal Texture Reconstruction for Dynamic Objects Using a Single RGB‐D Camera

DeepSurfels: Learning Online Appearance Fusion

Contact Info

Product

Resources

About