Computer Graphics Forum最新文献

The demand for immersive technology and realistic virtual environments has created a need for automated solutions to generate characters with morphological variations. However, existing approaches either rely on manual labour or oversimplify the problem by limiting it to static meshes or deformation transfers without shape morphing. In this paper, we propose a new cross-parameterization approach that semi-automates the generation of morphologically diverse characters with synthesized articulations and animations. The main contribution of this work is that our approach parameterizes deforming characters into a novel hierarchical multi-sphere domain, while considering the attributes of mesh topology, deformation and animation. With such a multi-sphere domain, our approach minimizes parametric distortion rates, enhances the bijectivity of parameterization and aligns deforming feature correspondences. The alignment process we propose allows users to focus only on major joint pairs, which is much simpler and more intuitive than the existing alignment solutions that involve a manual process of identifying feature points on mesh surfaces. Compared to recent works, our approach achieves high-quality results in the applications of 3D morphing, texture transfer, character synthesis and deformation transfer.

Hand in hand with the rapid development of machine learning, deep learning and generative AI algorithms and architectures, the graphics community has seen a remarkable evolution of novel techniques for material and appearance capture. Typically, these machine-learning-driven methods and technologies, in contrast to traditional techniques, rely on only a single or very few input images, while enabling the recovery of detailed, high-quality measurements of bi-directional reflectance distribution functions, as well as the corresponding spatially varying material properties, also known as Spatially Varying Bi-directional Reflectance Distribution Functions (SVBRDFs). Learning-based approaches for appearance capture will play a key role in the development of new technologies that will exhibit a significant impact on virtually all domains of graphics. Therefore, to facilitate future research, this State-of-the-Art Report (STAR) presents an in-depth overview of the state-of-the-art in machine-learning-driven material capture in general, and focuses on SVBRDF acquisition in particular, due to its importance in accurately modelling complex light interaction properties of real-world materials. The overview includes a categorization of current methods along with a summary of each technique, an evaluation of their functionalities, their complexity in terms of acquisition requirements, computational aspects and usability constraints. The STAR is concluded by looking forward and summarizing open challenges in research and development toward predictive and general appearance capture in this field. A complete list of the methods and papers reviewed in this survey is available at computergraphics.on.liu.se/star_svbrdf_dl/.

Experimental economics is an important branch of economics to study human behaviours in a controlled laboratory setting or out in the field. Scientific experiments are conducted in experimental economics to collect what decisions people make in specific circumstances and verify economic theories. As a significant couple of variables in the virtual experimental environment, decisions and outcomes change with the subjective factors of participants and objective circumstances, making it a difficult task to capture human behaviour patterns and establish correlations to verify economic theories. In this paper, we present a visual analytics system, EBPVis, which enables economists to visually explore human behaviour patterns and faithfully verify economic theories, e.g. the vicious cycle of poverty and poverty trap. We utilize a Doc2Vec model to transform the economic behaviours of participants into a vectorized space according to their sequential decisions, where frequent sequences can be easily perceived and extracted to represent human behaviour patterns. To explore the correlation between decisions and outcomes, an Outcome View is designed to display the outcome variables for behaviour patterns. We also provide a Comparison View to support an efficient comparison between multiple behaviour patterns by revealing their differences in terms of decision combinations and time-varying profits. Moreover, an Individual View is designed to illustrate the outcome accumulation and behaviour patterns of subjects. Case studies, expert feedback and user studies based on a real-world dataset have demonstrated the effectiveness and practicability of EBPVis in the representation of economic behaviour patterns and certification of economic theories.

Mixing textures is a basic and ubiquitous operation in data-driven algorithms for real-time texture generation and rendering. It is usually performed either by linear blending, or by cutting. We propose a new mixing operator which encompasses and extends both, creating more complex transitions that adapt to the texture's contents. Our mixing operator takes as input two or more textures along with two or more priority maps, which encode how the texture patterns should interact. The resulting mixed texture is defined pixel-wise by selecting the maximum of both priorities. We show that it integrates smoothly into two widespread applications: transition between two different textures, and texture synthesis that mixes pieces of the same texture. We provide constant-time and parallel evaluation of the resulting mix over square footprints of MIP-maps, making our operator suitable for real-time rendering. We also develop a micro-priority model, inspired by micro-geometry models in rendering, which represents sub-pixel priorities by a statistical distribution, and which allows for tuning between sharp cuts and smooth blend.

This paper presents an efficient algorithm for voxelizing the surface of triangular meshes in a single compute pass. The algorithm uses parallel equidistant lines to traverse the interior of triangles, minimizing costly memory operations and avoiding visiting the same voxels multiple times. By detecting and visiting only the voxels in each line operation, the proposed method achieves better performance results. This method incorporates a gap detection step, targeting areas where scanline-based voxelization methods might fail. By selectively addressing these gaps, our method attains superior performance outcomes. Additionally, the algorithm is written entirely in a single compute GLSL shader, which makes it highly portable and vendor independent. Its simplicity also makes it easy to adapt and extend for various applications. The paper compares the results of this algorithm with other modern methods, comprehensibly comparing the time performance and resources used. Additionally, we introduce a novel metric, the ‘Slope Consistency Value’, which quantifies triangle orientation's impact on voxelization accuracy for scanline-based approaches. The results show that the proposed solution outperforms existing, modern ones and obtains better results, especially in densely populated scenes with homogeneous triangle sizes and at higher resolutions.

Text-based hair design (TBHD) represents an innovative approach that utilizes text instructions for crafting hairstyle and colour, renowned for its flexibility and scalability. However, enhancing TBHD algorithms to improve generation quality and editing accuracy remains a current research difficulty. One important reason is that existing models fall short in alignment and fusion designs. Therefore, we propose a new layered multimodal fusion network called ETBHD-HMF, which decouples the input image and hair text information into layered hair colour and hairstyle representations. Within this network, the channel enhancement separation (CES) module is proposed to enhance important signals and suppress noise for text representation obtained from CLIP, thus improving generation quality. Based on this, we develop the weighted mapping fusion (WMF) sub-networks for hair colour and hairstyle. This sub-network applies the mapper operations to input image and text representations, acquiring joint information. The WMF then selectively merges image representation and joint information from various style layers using weighted operations, ultimately achieving fine-grained hairstyle designs. Additionally, to enhance editing accuracy and quality, we design a modality alignment loss to refine and optimize the information transmission and integration of the network. The experimental results of applying the network to the CelebA-HQ dataset demonstrate that our proposed model exhibits superior overall performance in terms of generation quality, visual realism, and editing accuracy. ETBHD-HMF (27.8 PSNR, 0.864 IDS) outperformed HairCLIP (26.9 PSNR, 0.828 IDS), with a 3% higher PSNR and a 4% higher IDS.

Texture editing is a crucial task in 3D modelling that allows users to automatically manipulate the surface materials of 3D models. However, the inherent complexity of 3D models and the ambiguous text description lead to the challenge of this task. To tackle this challenge, we propose ITEM3D, a Texture Editing Model designed for automatic 3D object editing according to the text Instructions. Leveraging the diffusion models and the differentiable rendering, ITEM3D takes the rendered images as the bridge between text and 3D representation and further optimizes the disentangled texture and environment map. Previous methods adopted the absolute editing direction, namely score distillation sampling (SDS) as the optimization objective, which unfortunately results in noisy appearances and text inconsistencies. To solve the problem caused by the ambiguous text, we introduce a relative editing direction, an optimization objective defined by the noise difference between the source and target texts, to release the semantic ambiguity between the texts and images. Additionally, we gradually adjust the direction during optimization to further address the unexpected deviation in the texture domain. Qualitative and quantitative experiments show that our ITEM3D outperforms the state-of-the-art methods on various 3D objects. We also perform text-guided relighting to show explicit control over lighting. Our project page: https://shengqiliu1.github.io/ITEM3D/.

U-Net structure is widely used for low-light image/video enhancement. The enhanced images result in areas with large local noise and loss of more details without proper guidance for global information. Attention mechanisms can better focus on and use global information. However, attention to images could significantly increase the number of parameters and computations. We propose a Row–Column Separated Attention module (RCSA) inserted after an improved U-Net. The RCSA module's input is the mean and maximum of the row and column of the feature map, which utilizes global information to guide local information with fewer parameters. We propose two temporal loss functions to apply the method to low-light video enhancement and maintain temporal consistency. Extensive experiments on the LOL, MIT Adobe FiveK image, and SDSD video datasets demonstrate the effectiveness of our approach.

3D point clouds stand as one of the prevalent representations for 3D data, offering the advantage of closely aligning with sensing technologies and providing an unbiased representation of a measured physical scene. Progressive compression is required for real-world applications operating on networked infrastructures with restricted or variable bandwidth. We contribute a novel approach that leverages a recursive binary space partition, where the partitioning planes are not necessarily axis-aligned and optimized via an entropy criterion. The planes are encoded via a novel adaptive quantization method combined with prediction. The input 3D point cloud is encoded as an interlaced stream of partitioning planes and number of points in the cells of the partition. Compared to previous work, the added value is an improved rate-distortion performance, especially for very low bitrates. The latter are critical for interactive navigation of large 3D point clouds on heterogeneous networked infrastructures.

Massachusetts Institute of Technology, Cambridge, MA, USA

June 24 – 26, 2024

Conference Co-Chairs

Justin Solomon, MIT

Mina Konaković Luković, MIT

Technical Program Co-Chairs

Ruizhen Hu, Shenzhen University

Sylvain Lefebvre, INRIA

Graduate School Co-Chairs

Silvia Sellán, University of Toronto

Edward Chien, Boston University

Steering Committee

Leif Kobbelt, RWTH Aachen University, DE

Marc Alexa, Technische Universität Berlin, DE

Pierre Alliez, INRIA, FR

Mirela Ben-Chen, Technion-IIT, IL

Hui Huang, Shenzhen University, CN

Niloy Mitra, University College London, GB

Daniele Panozzo, New York University, US

Alexa, Marc

TU Berlin, DE

Alliez, Pierre

Inria Sophia Antipolis, FR

Bærentzen, Jakob Andreas

Technical University of Denmark, DK

Belyaev, Alexander

Heriot-Watt University, GB

Ben-Chen, Mirela

Technion - Israel Institute of Technology, IL

Benes, Bedrich

Purdue University, US

Bommes, David

University of Bern, CH

Bonneel, Nicolas

CNRS, Université Lyon, FR

Botsch, Mario

TU Dortmund, DE

Boubekeur, Tamy

Adobe Research, FR

Campen, Marcel

Osnabrück University, DE

Chaine, Raphaelle

LIRIS CNRS, Université Lyon 1, FR

Chen, Renjie

University of Science and Technology of China, CN

Chen, Zhonggui

Xiamen University, CN

Chien, Edward

Boston University, US

Cignoni, Paolo

ISTI - CNR, IT

Cohen-Steiner, David

INRIA, FR

Desbrun, Mathieu

Inria / Ecole Polytechnique, FR

Dey, Tamal

Purdue University, US

Digne, Julie

LIRIS - CNRS, FR

Fu, Xiao-Ming

USTC, CN

Gao, Xifeng

Tencent America, US

Gingold, Yotam

George Mason University, US

Giorgi, Daniela

National Research Council of Italy, IT

Guerrero, Paul

Adobe Research, US

Herholz, Philipp

ETH Zurich, CH

Hildebrandt, Klaus

TU Delft, NL

Hoppe, Hugues

Independent Researcher, US

Hormann, Kai

Università della Svizzera italiana, CH

Huang, Jin

Zhejiang University, CN

Huang, Qixing

The University of Texas at Austin, US

Jacobson, Alec

University of Toronto and Adobe Research, CA

Ju, Tao

Washington University in St. Louis, US

Kazhdan, Misha

Johns Hopkins University, US

Keyser, John

Texas A & M University, US

Kim, Vladimir

Adobe, US

Kobbelt, Leif

RWTH Aachen University, DE

Kosinka, Jiri

Bern