가닥 지오메트리 이미지를 이용한 ConvNet 기반의 헤어 및 털 합성기
Hair and Fur Synthesizer via ConvNet Using Strand Geometry Images
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
본 논문에서는 라인 형태인 가닥(Strand) 지오메트리 이미지와 합성곱 신경망(Convolutional Neural Network, ConvNet 혹은 CNN)을 이용하여 저해상도 헤어 및 털 시뮬레이션을 고해상도로노이즈 없이 표현할 수 있는 기법을 제안한다. 저해상도와 고해상도 데이터 간의 쌍은 물리 기반시뮬레이션을 통해 얻을 수 있으며, 이렇게 얻어진 데이터를 이용하여 저해상도-고해상도 데이터쌍을 설정한다. 학습할 때 사용되는 데이터는 헤어 가닥 형태의 위치를 지오메트리 이미지로 변환하여 사용한다. 본 논문에서 제안하는 헤어 및 털 네트워크는 저해상도 이미지를 고해상도 이미지로 업스케일링(Upscaling)시키는 이미지 합성기를 위해 사용된다. 테스트 결과로 얻어진 고해상도 지오메트리 이미지가 고해상도 헤어로 다시 변환되면, 하나의 매핑 함수로 표현하기 어려운헤어의 찰랑거리는(Elastic) 움직임을 잘 표현할 수 있다. 합성 결과에 대한 성능으로 이전 물리 기반 시뮬레이션보다 빠른 성능을 보였으며, 복잡한 수치해석을 몰라도 쉽게 실행이 가능하다.
Abstract
In this paper, we propose a technique that can express low-resolution hair and fur simulations in high-resolution without noise using ConvNet and geometric images of strands in the form of lines. Pairs between low-resolution and high-resolution data can be obtained through physics-based simulation, and a low-resolution-high-resolution data pair is established using the obtained data. The data used for training is used by converting the position of the hair strands into a geometric image. The hair and fur network proposed in this paper is used for an image synthesizer that upscales a low-resolution image to a high-resolution image. If the high-resolution geometry image obtained as a result of the test is converted back to high-resolution hair, it is possible to express the elastic movement of hair, which is difficult to express with a single mapping function. As for the performance of the synthesis result, it showed faster performance than the traditional physics-based simulation, and it can be easily executed without knowing complex numerical analysis.
참고문헌 (38)
- Xiao, Xiao, "A physics-based virtual reality simulation framework for neonatal endotracheal intubation", 2020 IEEE Conference on Virtual Reality and 3D User Interfaces (VR), pp.557~565 (2020)
- Lee, Minjae, "A skinned tetrahedral mesh for hair animation and hair-water interaction", IEEE transactions on visualization and computer graphics, 25(3), pp.1449~1459 (2018)[DOI]
- Nealen, Andrew, "Physically based deformable models in computer graphics", Computer graphics forum, 25(4), pp.809~836 (2006), Blackwell Publishing Ltd
- Selle, Andrew, "A mass spring model for hair simulation", ACM SIGGRAPH 2008papers, pp.1~11 (2008)
- Müller, Matthias, "Position based dynamics", Journal of Visual Communication and Image Representation, 18(2), pp.109~118 (2007)[DOI]
- Müller, Matthias, "Fast Simulation of Inextensible Hair and Fur", VRIPHYS, 12, pp.39~44 (2012)[DOI]
- Bertails, Florence, "Super-helices for predicting the dynamics of natural hair", ACM Transactions on Graphics (TOG), 25(3), pp.1180~1187 (2006)[DOI]
- Hadap, Sunil, "Modeling dynamic hair as a continuum", Computer Graphics Forum, 20(3), pp.329~338 (2001), Blackwell Publishers Ltd
- Chai, Menglei, "Adaptive skinning for interactive hair-solid simulation", IEEE transactions on visualization and computer graphics, 23(7), pp.1725~1738 (2016)[DOI]
- Chai, Menglei, "Dynamic hair manipulation in images and videos", ACM Transactions on Graphics (TOG), 32(4), pp.1~8 (2013)[DOI]
- Ward, Kelly, "A survey on hair modeling: Styling, simulation, and rendering", IEEE transactions on visualization and computer graphics, 13(2), pp.213~234 (2007)[DOI]
- Battaglia, Peter W., "Relational inductive biases, deep learning, and graph networks", ~ (2018)
- Bergou, Miklós, "Discrete elastic rods", ACM SIGGRAPH 2008 papers, pp.1~12 (2008)
- Bertails, Florence, "Linear time super‐helices", Computer graphics forum, 28(2), pp.417~426 (2009), Blackwell Publishing Ltd
- Kubiak, Blazej, "A robust method for real-time thread simulation", Proceedings of the 2007 ACM symposium on Virtual reality software and technology, pp.85~88 (2007)
- Bergou, Miklós, "Discrete viscous threads", ACM Transactions on graphics (TOG), 29(4), pp.1~10 (2010)[DOI]
- Chang, Johnny T., "A practical model for hair mutual interactions", Proceedings of the 2002 ACM SIGGRAPH, 2002, 73~80
- Daldegan, Agnes, "An integrated system for modeling, animating and rendering hair", Computer Graphics Forum, 12(3), pp.211~221 (1993), Blackwell Science Ltd
- Kurihara, Tsuneya, "Models and Techniques in Computer Animation", pp.128~138 (1993), Springer
- Iben, Hayley, "Artistic simulation of curly hair", Proceedings of the 12th ACM SIGGRAPH, 2013, 63~71
- Yuksel, Cem, "Hair meshes", ACM Transactions on Graphics (TOG), 28(5), pp.1~7 (2009)[DOI]
- Ladický, L'ubor, "Data-driven fluid simulations using regression forests", ACM Transactions on Graphics (TOG), 34(6), pp.1~9 (2015)[DOI]
- Yang, Cheng, "Data‐driven projection method in fluid simulation", Computer Animation and Virtual Worlds, 27(3-4), pp.415~424 (2016)[DOI]
- Tompson, Jonathan, "Accelerating eulerian fluid simulation with convolutional networks", International Conference on Machine Learning, pp.3424~3433 (2017), PMLR
- Chu, Mengyu, "Data-driven synthesis of smoke flows with CNN-based feature descriptors", ACM Transactions on Graphics (TOG), 36(4), pp.1~14 (2017)[DOI]
- Um, Kiwon, "Liquid splash modeling with neural networks", Computer Graphics Forum, 37(8), pp.171~182 (2018)
- Wu, Jiajun, "Galileo: Perceiving physical object properties by integrating a physics engine with deep learning", Advances in neural information processing systems, 28 (2015)
- Fragkiadaki, Katerina, "Learning visual predictive models of physics for playing billiards", ~ (2015)
- Mottaghi, Roozbeh, "“What happens if...” Learning to Predict the Effect of Forces in Images", European conference on computer vision, pp.269~285 (2016), Springer
- Ummenhofer, Benjamin, "Lagrangian fluid simulation with continuous convolutions", International Conference on Learning Representations (2019)
- Scarselli, Franco, "The graph neural network model", IEEE transactions on neural networks, 20(1), pp.61~80 (2008)[DOI]
- Li, Yujia, "Gated graph sequence neural networks", ~ (2015)
- Gilmer, Justin, "Neural message passing for quantum chemistry", International conference on machine learning, pp.1263~1272 (2017), PMLR
- Battaglia, Peter, "Interaction networks for learning about objects, relations and physics", Advances in neural information processing systems, 29 (2016)
- Li, Yunzhu, "Learning particle dynamics for manipulating rigid bodies, deformable objects, and fluids", ~ (2018)
- Sanchez-Gonzalez, Alvaro, "Graph networks as learnable physics engines for inference and control", International Conference on Machine Learning, pp.4470~4479 (2018), PMLR
- Ba, Jimmy Lei, "Layer normalization", ~ (2016)
- Dong, Chao, "Image super-resolution using deep convolutional networks", IEEE transactions on pattern analysis and machine intelligence, 38(2), pp.295~307 (2015)[DOI]