SIGGRAPH 2017 论文列表

To date, material modeling in physically based computer animation has largely focused on mass and stiffness material properties. However, deformation dynamics is largely affected also by the damping properties. In this paper, we propose an interactive design method for nonlinear isotropic and anisotropic damping of complex three-dimensional solids simulated using the Finite Element Method (FEM). We first give a damping design method and interface whereby the user can set the damping properties so that motion aligned with each of a few chosen example deformations is damped by an independently prescribed amount, whereas the rest of the deformation space follows standard Rayleigh damping, or any viscous damping. Next, we demonstrate how to design nonlinear damping that depends on the magnitude of the deformation along each example deformation, by editing a single spline curve for each example deformation. Our user interface enables an art-directed and intuitive approach to controlling damping in solid simulations. We mathematically prove that our nonlinear anisotropic damping generalizes the frequency-dependent Caughey damping model, when starting from the Rayleigh damping. Finally, we give an inverse design method whereby the damping curve parameters can be inferred automatically from high-level user input, such as the amount of amplitude loss in one oscillation cycle along each of the chosen example deformations. To minimize numerical damping for implicit integration, we introduce an accurate and stable implicit integrator, which removes spurious high-frequency oscillations while only introducing a minimal amount of numerical damping. Our damping can generate effects not possible with previous methods, such as controllable nonlinear decaying envelopes whereby large deformations are damped faster or slower than small deformations, and damping anisotropic effects. We also fit our damping to videos of real-world objects undergoing large deformations, capturing their nonlinear and anisotropic damping dynamics.

Consumer time-of-flight depth cameras like Kinect and PMD are cheap, compact and produce video-rate depth maps in short-range applications. In this paper we apply energy-efficient epipolar imaging to the ToF domain to significantly expand the versatility of these sensors: we demonstrate live 3D imaging at over 15 m range outdoors in bright sunlight; robustness to global transport effects such as specular and diffuse inter-reflections---the first live demonstration for this ToF technology; interference-free 3D imaging in the presence of many ToF sensors, even when they are all operating at the same optical wavelength and modulation frequency; and blur-free, distortion-free 3D video in the presence of severe camera shake. We believe these achievements can make such cheap ToF devices broadly applicable in consumer and robotics domains.

We present a novel algorithm to control the physically-based animation of smoke. Given a set of keyframe smoke shapes, we compute a dense sequence of control force fields that can drive the smoke shape to match several keyframes at certain time instances. Our approach formulates this control problem as a spacetime optimization constrained by partial differential equations. In order to compute the locally optimal control forces, we alternatively optimize the velocity fields and density fields using an alternating direction method of multiplier (ADMM) optimizer. In order to reduce the high complexity of multiple passes of fluid resimulation during velocity field optimization, we utilize the coherence between consecutive fluid simulation passes. We demonstrate the benefits of our approach by computing accurate solutions on 2D and 3D benchmarks. In practice, we observe up to an order of magnitude improvement over prior optimal control methods.

Determining effective control strategies and solutions for high-degree-of-freedom humanoid characters has been a difficult, ongoing problem. A controller is only valid for a subset of the states of the character, known as the domain of attraction (DOA). This article shows how many states that are initially outside the DOA can be brought inside it. Our first contribution is to show how DOA expansion can be performed for a high-dimensional simulated character. Our second contribution is to present an algorithm that efficiently increases the DOA using random trees that provide denser coverage than the trees produced by typical sampling-based motion-planning algorithms. The trees are constructed offline but can be queried fast enough for near-real-time control. We show the effect of DOA expansion on getting up, crouch-to-stand, jumping, and standing-twist controllers. We also show how DOA expansion can be used to connect controllers together.

This paper addresses the problem of offline path and movement planning for wall climbing humanoid agents. We focus on simulating bouldering, i.e. climbing short routes with diverse moves, although we also demonstrate our system on a longer wall. Our approach combines a graph-based high-level path planner with low-level sampling-based optimization of climbing moves. Although the planning problem is complex, our system produces plausible solutions to bouldering problems (short climbing routes) in less than a minute. We further utilize a k-shortest paths approach, which enables the system to discover alternative paths - in climbing, alternative strategies often exist, and what might be optimal for one climber could be impossible for others due to individual differences in strength, flexibility, and reach. We envision our system could be used, e.g. in learning a climbing strategy, or as a test and evaluation tool for climbing route designers. To the best of our knowledge, this is the first paper to solve and simulate rich humanoid wall climbing, where more than one limb can move at the same time, and limbs can also hang free for balance or use wall friction in addition to predefined holds.

The paper presents a novel three-dimensional shape acquisition and reconstruction method based on the well-known Archimedes equality between fluid displacement and the submerged volume. By repeatedly dipping a shape in liquid in different orientations and measuring its volume displacement, we generate the dip transform : a novel volumetric shape representation that characterizes the object's surface. The key feature of our method is that it employs fluid displacements as the shape sensor. Unlike optical sensors, the liquid has no line-of-sight requirements, it penetrates cavities and hidden parts of the object, as well as transparent and glossy materials, thus bypassing all visibility and optical limitations of conventional scanning devices. Our new scanning approach is implemented using a dipping robot arm and a bath of water, via which it measures the water elevation. We show results of reconstructing complex 3D shapes and evaluate the quality of the reconstruction with respect to the number of dips.

We study the design and optimization of statically sound and materially efficient space structures constructed by connected beams. We propose a systematic computational framework for the design of space structures that incorporates static soundness, approximation of reference surfaces, boundary alignment, and geometric regularity. To tackle this challenging problem, we first jointly optimize node positions and connectivity through a nonlinear continuous optimization algorithm. Next, with fixed nodes and connectivity, we formulate the assignment of beam cross sections as a mixed-integer programming problem with a bilinear objective function and quadratic constraints. We solve this problem with a novel and practical alternating direction method based on linear programming relaxation. The capability and efficiency of the algorithms and the computational framework are validated by a variety of examples and comparisons.

Many approaches to shape comparison and recognition start by establishing a shape correspondence. We "turn the table" and show that quality shape correspondences can be obtained by performing many shape recognition tasks. What is more, the method we develop computes a fine-grained, topology-varying part correspondence between two 3D shapes where the core evaluation mechanism only recognizes shapes globally. This is made possible by casting the part correspondence problem in a deformation-driven framework and relying on a data-driven "deformation energy" which rates visual similarity between deformed shapes and models from a shape repository. Our basic premise is that if a correspondence between two chairs (or airplanes, bicycles, etc.) is correct, then a reasonable deformation between the two chairs anchored on the correspondence ought to produce plausible , "chair-like" in-between shapes. Given two 3D shapes belonging to the same category, we perform a top-down, hierarchical search for part correspondences. For a candidate correspondence at each level of the search hierarchy, we deform one input shape into the other, while respecting the correspondence, and rate the correspondence based on how well the resulting deformed shapes resemble other shapes from ShapeNet belonging to the same category as the inputs. The resemblance, i.e., plausibility, is measured by comparing multi-view depth images over category-specific features learned for the various shape categories. We demonstrate clear improvements over state-of-the-art approaches through tests covering extensive sets of man-made models with rich geometric and topological variations.

Learning physics-based locomotion skills is a difficult problem, leading to solutions that typically exploit prior knowledge of various forms. In this paper we aim to learn a variety of environment-aware locomotion skills with a limited amount of prior knowledge. We adopt a two-level hierarchical control framework. First, low-level controllers are learned that operate at a fine timescale and which achieve robust walking gaits that satisfy stepping-target and style objectives. Second, high-level controllers are then learned which plan at the timescale of steps by invoking desired step targets for the low-level controller. The high-level controller makes decisions directly based on high-dimensional inputs, including terrain maps or other suitable representations of the surroundings. Both levels of the control policy are trained using deep reinforcement learning. Results are demonstrated on a simulated 3D biped. Low-level controllers are learned for a variety of motion styles and demonstrate robustness with respect to force-based disturbances, terrain variations, and style interpolation. High-level controllers are demonstrated that are capable of following trails through terrains, dribbling a soccer ball towards a target location, and navigating through static or dynamic obstacles.

Extraction of structural lines from pattern-rich manga is a crucial step for migrating legacy manga to digital domain. Unfortunately, it is very challenging to distinguish structural lines from arbitrary, highly-structured, and black-and-white screen patterns. In this paper, we present a novel data-driven approach to identify structural lines out of pattern-rich manga, with no assumption on the patterns. The method is based on convolutional neural networks. To suit our purpose, we propose a deep network model to handle the large variety of screen patterns and raise output accuracy. We also develop an efficient and effective way to generate a rich set of training data pairs. Our method suppresses arbitrary screen patterns no matter whether these patterns are regular, irregular, tone-varying, or even pictorial, and regardless of their scales. It outputs clear and smooth structural lines even if these lines are contaminated by and immersed in complex patterns. We have evaluated our method on a large number of mangas of various drawing styles. Our method substantially outperforms state-of-the-art methods in terms of visual quality. We also demonstrate its potential in various manga applications, including manga colorization, manga retargeting, and 2.5D manga generation.

Example-based texture synthesis has been an active research problem for over two decades. Still, synthesizing textures with nonlocal structures remains a challenge. In this article, we present a texture synthesis technique that builds upon convolutional neural networks and extracted statistics of pretrained deep features. We introduce a structural energy, based on correlations among deep features, which capture the self-similarities and regularities characterizing the texture. Specifically, we show that our technique can synthesize textures that have structures of various scales, local and nonlocal, and the combination of the two.

Deep compositing is an important practical tool in creating digital imagery, but there has been little theoretical analysis of the underlying mathematical operators. Motivated by finding a simple formulation of the merging operation on OpenEXR -style deep images, we show that the Porter-Duff over function is the operator of a Lie group. In its corresponding Lie algebra, the splitting and mixing functions that OpenEXR deep merging requires have a particularly simple form. Working in the Lie algebra, we present a novel, simple proof of the uniqueness of the mixing function. The Lie group structure has many more applications, including new, correct resampling algorithms for volumetric images with alpha channels, and a deep image compression technique that outperforms that of OpenEXR .

Performance is a critical challenge in mobile image processing. Given a reference imaging pipeline, or even human-adjusted pairs of images, we seek to reproduce the enhancements and enable real-time evaluation. For this, we introduce a new neural network architecture inspired by bilateral grid processing and local affine color transforms. Using pairs of input/output images, we train a convolutional neural network to predict the coefficients of a locally-affine model in bilateral space. Our architecture learns to make local, global, and content-dependent decisions to approximate the desired image transformation. At runtime, the neural network consumes a low-resolution version of the input image, produces a set of affine transformations in bilateral space, upsamples those transformations in an edge-preserving fashion using a new slicing node, and then applies those upsampled transformations to the full-resolution image. Our algorithm processes high-resolution images on a smartphone in milliseconds, provides a real-time viewfinder at 1080p resolution, and matches the quality of state-of-the-art approximation techniques on a large class of image operators. Unlike previous work, our model is trained off-line from data and therefore does not require access to the original operator at runtime. This allows our model to learn complex, scene-dependent transformations for which no reference implementation is available, such as the photographic edits of a human retoucher.

In digital image editing software, layers organize images. However, layers are often not explicitly represented in the final image, and may never have existed for a scanned physical painting or a photograph. We propose a technique to decompose an image into layers. In our decomposition, each layer represents a single-color coat of paint applied with varying opacity. Our decomposition is based on the image’s RGB-space geometry. In RGB-space, the linear nature of the standard Porter-Duff [1984] “over” pixel compositing operation implies a geometric structure. The vertices of the convex hull of image pixels in RGB-space correspond to a palette of paint colors. These colors may be “hidden” and inaccessible to algorithms based on clustering visible colors. For our layer decomposition, users choose the palette size (degree of simplification to perform on the convex hull), as well as a layer order for the paint colors (vertices). We then solve a constrained optimization problem to find translucent, spatially coherent opacity for each layer, such that the composition of the layers reproduces the original image. We demonstrate the utility of the resulting decompositions for recoloring (global and local) and object insertion. Our layers can be interpreted as generalized barycentric coordinates; we compare to these and other recoloring approaches.

We present a novel data-driven algorithm to synthesize high resolution flow simulations with reusable repositories of space-time flow data. In our work, we employ a descriptor learning approach to encode the similarity between fluid regions with differences in resolution and numerical viscosity. We use convolutional neural networks to generate the descriptors from fluid data such as smoke density and flow velocity. At the same time, we present a deformation limiting patch advection method which allows us to robustly track deformable fluid regions. With the help of this patch advection, we generate stable space-time data sets from detailed fluids for our repositories. We can then use our learned descriptors to quickly localize a suitable data set when running a new simulation. This makes our approach very efficient, and resolution independent. We will demonstrate with several examples that our method yields volumes with very high effective resolutions, and non-dissipative small scale details that naturally integrate into the motions of the underlying flow.