Projects on 一橋大学ソーシャル・データサイエンス学部・研究科グラフィクス&ビジョン研究室 (谷田川達也研究室)

Bin-scanning: Segmentation of X-ray CT volume of binned parts using Morse skeleton graph of distance transform

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{yamauchi2023binscan,
 title = {Bin-scanning: Segmentation of X-ray CT volume of binned parts using Morse skeleton graph of distance transform},
 author = {Yamauchi, Yuta and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 journal = {Computational Visual Media},
 volume = {9},
 pages = {319--333},
 year = {2023},
 doi = {10.1007/s41095-022-0296-2}
}

Data Compression for Measured Heterogeneous Subsurface Scattering via Scattering Profile Blending

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article {yatagawa19data,
 title = {Data compression for measured subsurfce scattering via scattering profile blending},
 author = {Yatagawa, Tatsuya and Todo, Hideki and Yamaguchi, Yasushi and Morishima, Shigeo},
 journal = {The Visual Computer},
 volume = {36},
 number = {3},
 pages = {541--558},
 year = {2020},
 doi = {10.1007/s00371-018-01626-x},
 publisher = {Springer}
}

Movie

Demos

You can click the image to start playing video.
* The following videos are rendered in advance rather than being rendered at run time using WebGL.

Art. Stone

Ours

Song et al. 2009

Chessboard

Ours

Song et al. 2009

Onyx

Ours

Song et al. 2009

Blue Wax

Ours

Song et al. 2009

Yellow Wax

Ours

Song et al. 2009

Jade

Ours

Song et al. 2009

Deep Mesh Prior: Unsupervised Mesh Restoration using Graph Convolutional Networks

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings{hattori2021dmp,
 title = {{Deep Mesh Prior: Unsupervised Mesh Restoration using Graph Convolutional Networks}},
 author = {Hattori, Shota and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 booktitle = {CVPR Workshop on Learning to Generate 3D Shapes and Scenes},
 year = {2021},
 url = {https://arxiv.org/abs/2107.02909}
}

Degree of local symmetry for geometry-aware selective part visualisation on CT volumes

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{yasunami2022degree,
 title={Degree of local symmetry for geometry-aware selective part visualisation on CT volumes},
 author={Yasunami, Nobumichi and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 journal={Nondestructive Testing and Evaluation},
 pages={1--13},
 year={2022},
 publisher={Taylor \& Francis},
 doi={10.1080/10589759.2022.2090556}
}```

Efficient Metropolis Path Sampling for Material Editing and Re-rendering

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings {yamaguchi18efficient,
 booktitle = {Pacific Graphics Short Papers},
 editor = {Fu, Hongbo and Ghosh, Abhijeet and Kopf, Johannes},
 title = {{Efficient Metropolis Path Sampling for Material Editing and Re-rendering}},
 author = {Yamaguchi, Tomoya and Yatagawa, Tatsuya and Morishima, Shigeo},
 year = {2018},
 publisher = {The Eurographics Association},
 ISBN = {978-3-03868-073-4},
 DOI = {10.2312/pg.20181271}
}

Results

Cornell box #1

Cornell box #2

Lamp

Living room #1

Living room #2

End-to-End Deep Learning for Reconstructing Segmented 3D CT Image ffrom Multi-Energy X-ray Projections

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings{siqi2023ctseg,
 author={Wang, Siqi and Yatagawa, Tatsuya and Ohtake, Yutaka and Aoki, Toru and Hotta, Jun},
 title={End-to-End Deep Learning for Reconstructing Segmented 3D CT Image from Multi-Energy X-ray Projections},
 booktitle={Proceeding of the IEEE/CVF International Conference on Computer Vision (ICCV) Workshop},
 pages={2574--2582},
 year={2023}
}

Poster

Event-based Camera Simulation using Monte Carlo Path Tracing with Adaptive Denoising

Mon, 01 Jan 0001 00:00:00 +0000

Movie

Slides

Citation

@inproceedings{tsuji2023evr,
 author = {Yuta Tsuji and Tatsuya Yatagawa and Hiroyuki Kubo and Shigeo Morishima},
 title = {Event-Based Camera Simulation Using Monte Carlo Path Tracing with Adaptive Denoising},
 doi = {10.1109/icip49359.2023.10222771},
 year = {2023},
 month = {10},
 booktitle = {Proceedings of the International Conference on Image Processing ({ICIP})}
}

Extended Differentiable Marching Cubes by Manifold-Preserving Shape Inflation

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings{itoh2021ddmc,
 title = {{Extended Differentiable Marching Cubes by Manifold-Preserving Shape Inflation}},
 author = {Itoh, Kiichi and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 booktitle = {Proceedings of the British Machine Vision Conference (BMVC)},
 year = {2021},
 url = {https://www.bmvc2021-virtualconference.com/conference/papers/paper_0334.html}
}

FSNet: An Identity-Aware Generative Model for Image-based Face Swapping

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings {natsume18fsnet,
 title = {FSNet: An Identity-Aware Generative Model for Image-based Face Swapping},
 author = {Natsume, Ryota and Yatagawa, Tatsuya and Morishima, Shigeo},
 booktitle = {Proc. of Asian Conference on Computer Vision (ACCV)},
 month = {Dec},
 publisher = {Springer},
 year = {2018}
}

Results - Click image to enlarge

Face swapping

Different Face Orientation

Different Illumination

Swap between Same Identity

Image Based Measurement of Individual Fiber Lengths for Randomly Oriented Short Fiber Composites

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{siqi2022fiber,
 author={Wang, Siqi and Yatagawa, Tatsuya and Suzuki, Hiromasa and Ohtake, Yutaka},
 title={Image Based Measurement of Individual Fiber Lengths for Randomly Oriented Short Fiber Composites},
 journal={Journal of Nondestructive Evaluation},
 number={41},
 volume={45},
 page={},
 doi={10.1007/s10921-022-00876-z},
 year={2022}
}

Learning Scatter Artifact Correction in Cone-Beam X-Ray CT Using Incomplete Projections with Beam Hole Array

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{hattori2024learning,
 title={Learning Scatter Artifact Correction in Cone-Beam X-Ray CT Using Incomplete Projections with Beam Hole Array},
 author={Hattori, Haruki and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 journal={Journal of Nondestructive Evaluation},
 volume={43},
 number={3},
 pages={99},
 year={2024},
 publisher={Springer},
 doi = {10.1007/s10921-024-01113-5},
}

Learning Self-prior for Mesh Denoising using Dual Graph Convolutional Networks

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings{hattori2022ddmp,
 title = {Learning Self-prior for Mesh Denoising using Dual Graph Convolutional Networks},
 author = {Hattori, Shota and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 booktitle = {In Proceedings of the European Conference on Computer Vision (ECCV)},
 year = {2022},
 doi = {10.1007/978-3-031-20062-5_21}
}

Learning Self-Prior for Mesh Inpainting Using Self-Supervised Graph Convolutional Networks

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{hattori2024semigcn,
 title = {Learning Self-Prior for Mesh Inpainting Using Self-Supervised Graph Convolutional Networks},
 author = {Hattori, Shota and Yatagawa, Tatsuya and Ohtake, Yutaka and Suzuki, Hiromasa},
 journal = {IEEE Transactions on Visualization and Computer Graphics},
 year = {2024},
 doi = {10.1109/TVCG.2024.3364365}
}

Learning-Based Spotlight Position Optimization for Non-Line-of-Sight Human Localization and Posture Classification

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@InProceedings{Chandran_2024_WACV,
 author = {Chandran, Sreenithy and Yatagawa, Tatsuya and Kubo, Hiroyuki and Jayasuriya, Suren},
 title = {Learning-Based Spotlight Position Optimization for Non-Line-of-Sight Human Localization and Posture Classification},
 booktitle = {Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
 month = {January},
 year = {2024},
 pages = {4218-4227}
}

Light Source Selection in Primary-Sample-Space Neural Photon Sampling

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings{tsuji2021nppm,
 author={Tsuji, Yuta and Yatagawa, Tatsuya and Morishima, Shigeo},
 title={Light Source Selection in Primary-Sample-Space Neural Photon Sampling},
 booktitle={SIGGRAPH Asia 2021 Posters},
 number={},
 volume={},
 articleno={38},
 page={1--2},
 doi={10.1145/3476124.3488639},
 year={2021}
}

Poster

LinSSS: Linear decomposition of heterogeneous subsurface scattering for real-time screen-space rendering

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{yatagawa2020linsss,
 title = {{LinSSS: Linear decomposition of heterogeneous subsurface scattering for real-time screen-space rendering}},
 author = {Yatagawa, Tatsuya and Yamaguchi, Yasushi and Morishima, Shigeo},
 journal = {The Visual Computer},
 volume = {36},
 number = {10},
 pages = {1979--1992},
 year = {2020},
 doi = {10.1007/s00371-020-01915-4}
}

Movie

Monte Carlo Path Tracing and Statistical Event Detection for Event Camera Simulation

Mon, 01 Jan 0001 00:00:00 +0000

Movie

Slides

Citation

@inproceedings{manabe2024evr,
 author = {Yuichiro Manabe and Tatsuya Yatagawa and Shigeo Morishima and Hiroyuki Kubo},
 title = {Monte Carlo Path Tracing and Statistical Event Detection for Event Camera Simulation},
 booktitle = {Proceedings of the International Conference on Computational Photography ({ICCP})},
 year = {2024},
 doi = {10.1109/ICCP61108.2024.10644728},
}

NeMCoF: Neural Material Composition Fields for Material Decomposition in Sparse-View Spectral X-ray CT

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{hotta2025nemcof,
 title={NeMCoF: Neural Material Composition Fields for Material Decomposition in Sparse-View Spectral X-ray CT},
 author={Hotta, Takumi and Yatagawa, Tatsuya and Ohtake, Yutaka and Aoki, Toru},
 journal={Journal of Nondestructive Evaluation},
 volume={44},
 number={4},
 pages={122},
 year={2025},
 publisher={Springer},
 doi={10.1007/s10921-025-01263-0},
}

Primitive-based Model Reduction for Huge 3D Meshes of Industrial Plants

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@inproceedings{hattori2024model,
 title = {Primitive-based Model Reduction for Huge 3D Meshes of Industrial Plants},
 author = {Hattori, Haruki and Ohtake, Yutaka and Yatagawa, Tatsuya and Miyamoto, Kazutoshi and Hirota, Masashi},
 booktitle = {IEEE Conference on Virtual Reality and 3D User Interfaces (IEEE VR)},
 year = {2024},
}

Poster

Real-time Indirect Illumination of Emissive Inhomogeneous Volumes using Layered Polygonal Area Lights

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@journal{kuge19lpal,
 title = {{Real-time Indirect Illumination of Emissive Inhomogeneous Volumes using Layered Polygonal Area Lights}},
 author = {Kuge, Takahiro and Yatagawa, Tatsuya and Morishima, Shigeo},
 journal = {Computer Graphics Forum},
 volume = {38},
 number = {7},
 pages = {449--460},
 year = {2019},
 doi = {10.1111/cgf.13851}
}

Movie

Real-time Rendering of Layered Materials with Anisotropic Normal Distributions

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{yamaguchi2020anisolayer,
 title = {{Real-time rendering of layered materials with anisotorpic normal distributions}},
 author = {Yamaguchi, Tomoya and Yatagawa, Tatsuya and Tokuyoshi, Yusuke and Morishima, Shigeo},
 journal = {Computational Visual Media},
 volume = {6},
 pages = {29--36},
 year = {2020},
 doi = {10.1007/s41095-019-0154-z}
}

@inproceedings{yamaguchi2019anisolayer,
 title = {{Real-time Rendering of Layered Materials with Anisotorpic Normal Distributions}},
 author = {Yamaguchi, Tomoya and Yatagawa, Tatsuya and Tokuyoshi, Yusuke and Morishima, Shigeo},
 booktitle = {SIGGRAPH Asia 2019, Technical Brief},
 year = {2019},
 doi = {10.1145/3355088.3365165},
 publisher = {ACM},
}

Movie

Real-time Shading with Free-form Planar Area Lights using Linearly Transformed Cosines

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{Kuge2022BezLight,
 author = {Takahiro Kuge and Tatsuya Yatagawa and Shigeo Morishima},
issn = {2331-7418},
 journal = {Journal of Computer Graphics Techniques (JCGT)},
 month = {February},
 day = {18},
 number = {1},
 pages = {1--16},
 title = {Real-time Shading with Free-form Planar Area Lights using Linearly Transformed Cosines},
 url = {http://jcgt.org/published/0011/01/01/},
 volume = {11},
 year = {2022},
}

Movie

Slides (at I3D 2022)

Poster (at I3D 2022)

RSGAN: Face Swapping and Editing using Face and Hair Representation in Latent Spaces

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{natsume18rsgan,
 title = {RSGAN: Face Swapping and Editing using Face and Hair Representation in Latent Spaces},
 author = {Natsume, Ryota and Yatagawa, Tatsuya and Morishima, Shigeo},
 journal = {arXiv preprint arXiv:1804.03447},
 year = {2018}
}

Results - Click image to enlarge

Face swap

Conditioned face swap

Random part sampling

Part interpolation

Sparse pixel sampling for appearance edit propagation

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article {yatagawa15sparse,
 title = {{Sparse pixel sampling for appearance edit propagation}},
 author = {Yatagawa, Tatsuya and Yamaguchi, Yasushi},
 journal = {The Visual Computer},
 year={2015},
 month={Jun},
 volume={31},
 number={6},
 pages={1101--1111},
 ISBN = {1432-2315},
 DOI = {10.1007/s00371-015-1094-y}
}

Image Editing

Video Editing

Input

Please use a browser which supports the HTML5 video tag.

Output

Please use a browser which supports the HTML5 video tag.

← push left button to play

Demo 1

Demo 2 (interactive mode)

MATLAB demo usage

A simple MATLAB implementation of the proposed system is provided in the supplemental materials. The demo program includes two main scripts:

Sparse-View Cone-Beam CT Reconstruction by Bar-by-Bar Neural FDK Algorithm

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{siqi2023bbb,
 author={Wang, Siqi and Yatagawa, Tatsuya and Suzuki, Hiromasa and Ohtake, Yutaka},
 title={Sparse-View Cone-Beam CT Reconstruction by Bar-by-Bar Neural FDK Algorithm},
 journal={Nondestructive Testing and Evaluation},
 number={},
 volume={},
 page={},
 doi={10.1080/10589759.2023.2195646},
 year={2023}
}

Template-Based Dynamic-Background Removal for Fast Video Completion

Mon, 01 Jan 0001 00:00:00 +0000

Abstract

Video completion is a technique to remove large portion of area in video by detecting other similar portion from input video. Recently, preliminarily classifying input video pixel into foreground and background is found to be efficient for accelerating video completion procedure. Though existing completion algorithms only pay attention to moving foreground, many videos contain moving background. Therefore they are originally difficult to handle videos with moving background. Many background subtraction algorithms can extract moving objects without any user guide, but they cannot distinguish moving foreground and moving background. We presents the novel approach to remove moving background region automatically and apply existing fast completion algorithms to videos with moving background.

Results

Foreground Extraction

(upper left) input video

(upper right) simple thresholding

(lower left) Zivkovic 2004 [1]

(lower right) our result

Completion 01

(left) input video

(right) output video

Completion 02

(left) input video

(right) output video

Temporally Coherent Video Editing using an Edit Propagation Matrix

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{yatagawa14efficient,
	journal = {Computers \& Graphics},
	title = {{Temporally coherent video editing using an edit propagation matrix}},
	author = {Yatagawa, Tatsuya and Yamaguchi, Yasushi},
	year = {2014},
	publisher = {Elsevier},
	volume = {43},
	pages = {1--10},
	ISSN = {0097-8493},
	DOI = {10.1016/j.cag.2014.05.001}
}

Movie

Towards accelerating polarization path tracing of multi-bounce Smith microfacet BSDFs

Mon, 01 Jan 0001 00:00:00 +0000

Poster

Citation

@inproceedings{ohba2025towards,
 author = {Ohba, Hidehito and Yatagawa, Tatsuya and Morishima, Shigeo},
 title = {Towards accelerating polarization path tracing of multi-bounce Smith microfacet BSDFs},
 year = {2025},
 publisher = {Association for Computing Machinery},
 doi = {10.1145/3721250.3742996},
 booktitle = {Proceedings of the Special Interest Group on Computer Graphics and Interactive Techniques Conference Posters},
 articleno = {76},
 numpages = {2},
 series = {SIGGRAPH Posters '25}
}

X-ray 3D Fiber Orientation Tomography via Alternating Optimization of Scattering Coefficients and Directions

Mon, 01 Jan 0001 00:00:00 +0000

Citation

@article{mori2024xray,
 title = {X-ray 3D Fiber Orientation Tomography via Alternating Optimization of Scattering Coefficients and Directions},
 author = {Mori, Tomoki and Ohtake, Yutaka and Yatagawa, Tatsuya and Kido, Kazuhiro and Tsuboi, Yasunori},
 journal = {Journal of Nondestructive Evaluation},
 volume = {43},
 number = {2},
 year = {2024},
 doi = {10.1007/s10921-024-01066-9}
}

Projects on 一橋大学 ソーシャル・データサイエンス学部・研究科 グラフィクス&ビジョン研究室 (谷田川達也研究室)

Bin-scanning: Segmentation of X-ray CT volume of binned parts using Morse skeleton graph of distance transform

Citation

Data Compression for Measured Heterogeneous Subsurface Scattering via Scattering Profile Blending

Citation

Movie

Demos

Art. Stone

Ours

Song et al. 2009

Chessboard

Ours

Song et al. 2009

Onyx

Ours

Song et al. 2009

Blue Wax

Ours

Song et al. 2009

Yellow Wax

Ours

Song et al. 2009

Jade

Ours

Song et al. 2009

Deep Mesh Prior: Unsupervised Mesh Restoration using Graph Convolutional Networks

Citation

Degree of local symmetry for geometry-aware selective part visualisation on CT volumes

Citation

Efficient Metropolis Path Sampling for Material Editing and Re-rendering

Citation

Results

Cornell box #1

Cornell box #2

Lamp

Living room #1

Living room #2

End-to-End Deep Learning for Reconstructing Segmented 3D CT Image ffrom Multi-Energy X-ray Projections

Citation

Poster

Event-based Camera Simulation using Monte Carlo Path Tracing with Adaptive Denoising

Movie

Slides

Citation

Extended Differentiable Marching Cubes by Manifold-Preserving Shape Inflation

Citation

FSNet: An Identity-Aware Generative Model for Image-based Face Swapping

Citation

Results - Click image to enlarge

Face swapping

Different Face Orientation

Different Illumination

Swap between Same Identity

Image Based Measurement of Individual Fiber Lengths for Randomly Oriented Short Fiber Composites

Citation

Learning Scatter Artifact Correction in Cone-Beam X-Ray CT Using Incomplete Projections with Beam Hole Array

Citation

Learning Self-prior for Mesh Denoising using Dual Graph Convolutional Networks

Citation

Learning Self-Prior for Mesh Inpainting Using Self-Supervised Graph Convolutional Networks

Citation

Learning-Based Spotlight Position Optimization for Non-Line-of-Sight Human Localization and Posture Classification

Citation

Light Source Selection in Primary-Sample-Space Neural Photon Sampling

Citation

Poster

LinSSS: Linear decomposition of heterogeneous subsurface scattering for real-time screen-space rendering

Citation

Movie

Monte Carlo Path Tracing and Statistical Event Detection for Event Camera Simulation

Movie

Slides

Citation

NeMCoF: Neural Material Composition Fields for Material Decomposition in Sparse-View Spectral X-ray CT

Citation

Primitive-based Model Reduction for Huge 3D Meshes of Industrial Plants

Citation

Poster

Real-time Indirect Illumination of Emissive Inhomogeneous Volumes using Layered Polygonal Area Lights

Citation

Projects on 一橋大学ソーシャル・データサイエンス学部・研究科グラフィクス&ビジョン研究室 (谷田川達也研究室)