Showing papers in "ACM Transactions on Graphics in 2019"

Dynamic Graph CNN for Learning on Point Clouds

Abstract: Point clouds provide a flexible geometric representation suitable for countless applications in computer graphics; they also comprise the raw output of most 3D data acquisition devices. While hand-designed features on point clouds have long been proposed in graphics and vision, however, the recent overwhelming success of convolutional neural networks (CNNs) for image analysis suggests the value of adapting insight from CNN to the point cloud world. Point clouds inherently lack topological information, so designing a model to recover topology can enrich the representation power of point clouds. To this end, we propose a new neural network module dubbed EdgeConv suitable for CNN-based high-level tasks on point clouds, including classification and segmentation. EdgeConv acts on graphs dynamically computed in each layer of the network. It is differentiable and can be plugged into existing architectures. Compared to existing modules operating in extrinsic space or treating each point independently, EdgeConv has several appealing properties: It incorporates local neighborhood information; it can be stacked applied to learn global shape properties; and in multi-layer systems affinity in feature space captures semantic characteristics over potentially long distances in the original embedding. We show the performance of our model on standard benchmarks, including ModelNet40, ShapeNetPart, and S3DIS.

Deferred neural rendering: image synthesis using neural textures

Abstract: The modern computer graphics pipeline can synthesize images at remarkable visual quality; however, it requires well-defined, high-quality 3D content as input. In this work, we explore the use of imperfect 3D content, for instance, obtained from photo-metric reconstructions with noisy and incomplete surface geometry, while still aiming to produce photo-realistic (re-)renderings. To address this challenging problem, we introduce Deferred Neural Rendering, a new paradigm for image synthesis that combines the traditional graphics pipeline with learnable components. Specifically, we propose Neural Textures, which are learned feature maps that are trained as part of the scene capture process. Similar to traditional textures, neural textures are stored as maps on top of 3D mesh proxies; however, the high-dimensional feature maps contain significantly more information, which can be interpreted by our new deferred neural rendering pipeline. Both neural textures and deferred neural renderer are trained end-to-end, enabling us to synthesize photo-realistic images even when the original 3D content was imperfect. In contrast to traditional, black-box 2D generative neural networks, our 3D representation gives us explicit control over the generated output, and allows for a wide range of application domains. For instance, we can synthesize temporally-consistent video re-renderings of recorded 3D scenes as our representation is inherently embedded in 3D space. This way, neural textures can be utilized to coherently re-render or manipulate existing video content in both static and dynamic environments at real-time rates. We show the effectiveness of our approach in several experiments on novel view synthesis, scene editing, and facial reenactment, and compare to state-of-the-art approaches that leverage the standard graphics pipeline as well as conventional generative neural networks.

MeshCNN: a network with an edge

Abstract: Polygonal meshes provide an efficient representation for 3D shapes. They explicitly captureboth shape surface and topology, and leverage non-uniformity to represent large flat regions as well as sharp, intricate features. This non-uniformity and irregularity, however, inhibits mesh analysis efforts using neural networks that combine convolution and pooling operations. In this paper, we utilize the unique properties of the mesh for a direct analysis of 3D shapes using MeshCNN, a convolutional neural network designed specifically for triangular meshes. Analogous to classic CNNs, MeshCNN combines specialized convolution and pooling layers that operate on the mesh edges, by leveraging their intrinsic geodesic connections. Convolutions are applied on edges and the four edges of their incident triangles, and pooling is applied via an edge collapse operation that retains surface topology, thereby, generating new mesh connectivity for the subsequent convolutions. MeshCNN learns which edges to collapse, thus forming a task-driven process where the network exposes and expands the important features while discarding the redundant ones. We demonstrate the effectiveness of MeshCNN on various learning tasks applied to 3D meshes.

Local light field fusion: practical view synthesis with prescriptive sampling guidelines

Abstract: We present a practical and robust deep learning solution for capturing and rendering novel views of complex real world scenes for virtual exploration. Previous approaches either require intractably dense view sampling or provide little to no guidance for how users should sample views of a scene to reliably render high-quality novel views. Instead, we propose an algorithm for view synthesis from an irregular grid of sampled views that first expands each sampled view into a local light field via a multiplane image (MPI) scene representation, then renders novel views by blending adjacent local light fields. We extend traditional plenoptic sampling theory to derive a bound that specifies precisely how densely users should sample views of a given scene when using our algorithm. In practice, we apply this bound to capture and render views of real world scenes that achieve the perceptual quality of Nyquist rate view sampling while using up to 4000X fewer views. We demonstrate our approach's practicality with an augmented reality smart-phone app that guides users to capture input images of a scene and viewers that enable realtime virtual exploration on desktop and mobile platforms.

Neural volumes: learning dynamic renderable volumes from images

Abstract: Modeling and rendering of dynamic scenes is challenging, as natural scenes often contain complex phenomena such as thin structures, evolving topology, translucency, scattering, occlusion, and biological motion. Mesh-based reconstruction and tracking often fail in these cases, and other approaches (e.g., light field video) typically rely on constrained viewing conditions, which limit interactivity. We circumvent these difficulties by presenting a learning-based approach to representing dynamic objects inspired by the integral projection model used in tomographic imaging. The approach is supervised directly from 2D images in a multi-view capture setting and does not require explicit reconstruction or tracking of the object. Our method has two primary components: an encoder-decoder network that transforms input images into a 3D volume representation, and a differentiable ray-marching operation that enables end-to-end training. By virtue of its 3D representation, our construction extrapolates better to novel viewpoints compared to screen-space rendering techniques. The encoder-decoder architecture learns a latent representation of a dynamic scene that enables us to produce novel content sequences not seen during training. To overcome memory limitations of voxel-based representations, we learn a dynamic irregular grid structure implemented with a warp field during ray-marching. This structure greatly improves the apparent resolution and reduces grid-like artifacts and jagged motion. Finally, we demonstrate how to incorporate surface-based representations into our volumetric-learning framework for applications where the highest resolution is required, using facial performance capture as a case in point.

Semantic photo manipulation with a generative image prior

Abstract: Despite the recent success of GANs in synthesizing images conditioned on inputs such as a user sketch, text, or semantic labels, manipulating the high-level attributes of an existing natural photograph with GANs is challenging for two reasons. First, it is hard for GANs to precisely reproduce an input image. Second, after manipulation, the newly synthesized pixels often do not fit the original image. In this paper, we address these issues by adapting the image prior learned by GANs to image statistics of an individual image. Our method can accurately reconstruct the input image and synthesize new content, consistent with the appearance of the input image. We demonstrate our interactive system on several semantic image editing tasks, including synthesizing new objects consistent with background, removing unwanted objects, and changing the appearance of an object. Quantitative and qualitative comparisons against several existing methods demonstrate the effectiveness of our method.

Mitsuba 2: a retargetable forward and inverse renderer

Abstract: Modern rendering systems are confronted with a dauntingly large and growing set of requirements: in their pursuit of realism, physically based techniques must increasingly account for intricate properties of light, such as its spectral composition or polarization. To reduce prohibitive rendering times, vectorized renderers exploit coherence via instruction-level parallelism on CPUs and GPUs. Differentiable rendering algorithms propagate derivatives through a simulation to optimize an objective function, e.g., to reconstruct a scene from reference images. Catering to such diverse use cases is challenging and has led to numerous purpose-built systems---partly, because retrofitting features of this complexity onto an existing renderer involves an error-prone and infeasibly intrusive transformation of elementary data structures, interfaces between components, and their implementations (in other words, everything). We propose Mitsuba 2, a versatile renderer that is intrinsically retargetable to various applications including the ones listed above. Mitsuba 2 is implemented in modern C++ and leverages template metaprogramming to replace types and instrument the control flow of components such as BSDFs, volumes, emitters, and rendering algorithms. At compile time, it automatically transforms arithmetic, data structures, and function dispatch, turning generic algorithms into a variety of efficient implementations without the tedium of manual redesign. Possible transformations include changing the representation of color, generating a "wide" renderer that operates on bundles of light paths, just-in-time compilation to create computational kernels that run on the GPU, and forward/reverse-mode automatic differentiation. Transformations can be chained, which further enriches the space of algorithms derived from a single generic implementation. We demonstrate the effectiveness and simplicity of our approach on several applications that would be very challenging to create without assistance: a rendering algorithm based on coherent MCMC exploration, a caustic design method for gradient-index optics, and a technique for reconstructing heterogeneous media in the presence of multiple scattering.

LiveCap: Real-Time Human Performance Capture From Monocular Video

Abstract: We present the first real-time human performance capture approach that reconstructs dense, space-time coherent deforming geometry of entire humans in general everyday clothing from just a single RGB video. We propose a novel two-stage analysis-by-synthesis optimization whose formulation and implementation are designed for high performance. In the first stage, a skinned template model is jointly fitted to background subtracted input video, 2D and 3D skeleton joint positions found using a deep neural network, and a set of sparse facial landmark detections. In the second stage, dense non-rigid 3D deformations of skin and even loose apparel are captured based on a novel real-time capable algorithm for non-rigid tracking using dense photometric and silhouette constraints. Our novel energy formulation leverages automatically identified material regions on the template to model the differing non-rigid deformation behavior of skin and apparel. The two resulting non-linear optimization problems per frame are solved with specially tailored data-parallel Gauss-Newton solvers. To achieve real-time performance of over 25Hz, we design a pipelined parallel architecture using the CPU and two commodity GPUs. Our method is the first real-time monocular approach for full-body performance capture. Our method yields comparable accuracy with off-line performance capture techniques while being orders of magnitude faster.

Learning to optimize halide with tree search and random programs

Abstract: We present a new algorithm to automatically schedule Halide programs for high-performance image processing and deep learning. We significantly improve upon the performance of previous methods, which considered a limited subset of schedules. We define a parameterization of possible schedules much larger than prior methods and use a variant of beam search to search over it. The search optimizes runtime predicted by a cost model based on a combination of new derived features and machine learning. We train the cost model by generating and featurizing hundreds of thousands of random programs and schedules. We show that this approach operates effectively with or without autotuning. It produces schedules which are on average almost twice as fast as the existing Halide autoscheduler without autotuning, or more than twice as fast with, and is the first automatic scheduling algorithm to significantly outperform human experts on average.

3D Ken Burns effect from a single image

Abstract: The Ken Burns effect allows animating still images with a virtual camera scan and zoom. Adding parallax, which results in the 3D Ken Burns effect, enables significantly more compelling results. Creating such effects manually is time-consuming and demands sophisticated editing skills. Existing automatic methods, however, require multiple input images from varying viewpoints. In this paper, we introduce a framework that synthesizes the 3D Ken Burns effect from a single image, supporting both a fully automatic mode and an interactive mode with the user controlling the camera. Our framework first leverages a depth prediction pipeline, which estimates scene depth that is suitable for view synthesis tasks. To address the limitations of existing depth estimation methods such as geometric distortions, semantic distortions, and inaccurate depth boundaries, we develop a semantic-aware neural network for depth prediction, couple its estimate with a segmentation-based depth adjustment process, and employ a refinement neural network that facilitates accurate depth predictions at object boundaries. According to this depth estimate, our framework then maps the input image to a point cloud and synthesizes the resulting video frames by rendering the point cloud from the corresponding camera positions. To address disocclusions while maintaining geometrically and temporally coherent synthesis results, we utilize context-aware color- and depth-inpainting to fill in the missing information in the extreme views of the camera path, thus extending the scene geometry of the point cloud. Experiments with a wide variety of image content show that our method enables realistic synthesis results. Our study demonstrates that our system allows users to achieve better results while requiring little effort compared to existing solutions for the 3D Ken Burns effect creation.

Single image portrait relighting

Abstract: Lighting plays a central role in conveying the essence and depth of the subject in a portrait photograph. Professional photographers will carefully control the lighting in their studio to manipulate the appearance of their subject, while consumer photographers are usually constrained to the illumination of their environment. Though prior works have explored techniques for relighting an image, their utility is usually limited due to requirements of specialized hardware, multiple images of the subject under controlled or known illuminations, or accurate models of geometry and reflectance. To this end, we present a system for portrait relighting: a neural network that takes as input a single RGB image of a portrait taken with a standard cellphone camera in an unconstrained environment, and from that image produces a relit image of that subject as though it were illuminated according to any provided environment map. Our method is trained on a small database of 18 individuals captured under different directional light sources in a controlled light stage setup consisting of a densely sampled sphere of lights. Our proposed technique produces quantitatively superior results on our dataset's validation set compared to prior works, and produces convincing qualitative relighting results on a dataset of hundreds of real-world cellphone portraits. Because our technique can produce a 640 × 640 image in only 160 milliseconds, it may enable interactive user-facing photographic applications in the future.

Neural Importance Sampling

Abstract: We propose to use deep neural networks for generating samples in Monte Carlo integration. Our work is based on non-linear independent components estimation (NICE), which we extend in numerous ways to improve performance and enable its application to integration problems. First, we introduce piecewise-polynomial coupling transforms that greatly increase the modeling power of individual coupling layers. Second, we propose to preprocess the inputs of neural networks using one-blob encoding, which stimulates localization of computation and improves inference. Third, we derive a gradient-descent-based optimization for the Kullback-Leibler and the χ2 divergence for the specific application of Monte Carlo integration with unnormalized stochastic estimates of the target distribution. Our approach enables fast and accurate inference and efficient sample generation independently of the dimensionality of the integration domain. We show its benefits on generating natural images and in two applications to light-transport simulation: first, we demonstrate learning of joint path-sampling densities in the primary sample space and importance sampling of multi-dimensional path prefixes thereof. Second, we use our technique to extract conditional directional densities driven by the product of incident illumination and the BSDF in the rendering equation, and we leverage the densities for path guiding. In all applications, our approach yields on-par or higher performance than competing techniques at equal sample count.

GRAINS: Generative Recursive Autoencoders for INdoor Scenes

Abstract: We present a generative neural network that enables us to generate plausible 3D indoor scenes in large quantities and varieties, easily and highly efficiently. Our key observation is that indoor scene structures are inherently hierarchical. Hence, our network is not convolutional; it is a recursive neural network, or RvNN. Using a dataset of annotated scene hierarchies, we train a variational recursive autoencoder, or RvNN-VAE, which performs scene object grouping during its encoding phase and scene generation during decoding. Specifically, a set of encoders are recursively applied to group 3D objects based on support, surround, and co-occurrence relations in a scene, encoding information about objects’ spatial properties, semantics, and relative positioning with respect to other objects in the hierarchy. By training a variational autoencoder (VAE), the resulting fixed-length codes roughly follow a Gaussian distribution. A novel 3D scene can be generated hierarchically by the decoder from a randomly sampled code from the learned distribution. We coin our method GRAINS, for Generative Recursive Autoencoders for INdoor Scenes. We demonstrate the capability of GRAINS to generate plausible and diverse 3D indoor scenes and compare with existing methods for 3D scene synthesis. We show applications of GRAINS including 3D scene modeling from 2D layouts, scene editing, and semantic scene segmentation via PointNet whose performance is boosted by the large quantity and variety of 3D scenes generated by our method.

SDM-NET: deep generative network for structured deformable mesh

Abstract: We introduce SDM-NET, a deep generative neural network which produces structured deformable meshes. Specifically, the network is trained to generate a spatial arrangement of closed, deformable mesh parts, which respects the global part structure of a shape collection, e.g., chairs, airplanes, etc. Our key observation is that while the overall structure of a 3D shape can be complex, the shape can usually be decomposed into a set of parts, each homeomorphic to a box, and the finer-scale geometry of the part can be recovered by deforming the box. The architecture of SDM-NET is that of a two-level variational autoencoder (VAE). At the part level, a PartVAE learns a deformable model of part geometries. At the structural level, we train a Structured Parts VAE (SP-VAE), which jointly learns the part structure of a shape collection and the part geometries, ensuring the coherence between global shape structure and surface details. Through extensive experiments and comparisons with the state-of-the-art deep generative models of shapes, we demonstrate the superiority of SDM-NET in generating meshes with visual quality, flexible topology, and meaningful structures, benefiting shape interpolation and other subsequent modeling tasks.

Scalable muscle-actuated human simulation and control

Abstract: Many anatomical factors, such as bone geometry and muscle condition, interact to affect human movements. This work aims to build a comprehensive musculoskeletal model and its control system that reproduces realistic human movements driven by muscle contraction dynamics. The variations in the anatomic model generate a spectrum of human movements ranging from typical to highly stylistic movements. To do so, we discuss scalable and reliable simulation of anatomical features, robust control of under-actuated dynamical systems based on deep reinforcement learning, and modeling of pose-dependent joint limits. The key technical contribution is a scalable, two-level imitation learning algorithm that can deal with a comprehensive full-body musculoskeletal model with 346 muscles. We demonstrate the predictive simulation of dynamic motor skills under anatomical conditions including bone deformity, muscle weakness, contracture, and the use of a prosthesis. We also simulate various pathological gaits and predictively visualize how orthopedic surgeries improve post-operative gaits.

StructureNet: hierarchical graph networks for 3D shape generation

Abstract: The ability to generate novel, diverse, and realistic 3D shapes along with associated part semantics and structure is central to many applications requiring high-quality 3D assets or large volumes of realistic training data. A key challenge towards this goal is how to accommodate diverse shape variations, including both continuous deformations of parts as well as structural or discrete alterations which add to, remove from, or modify the shape constituents and compositional structure. Such object structure can typically be organized into a hierarchy of constituent object parts and relationships, represented as a hierarchy of n-ary graphs. We introduce StructureNet, a hierarchical graph network which (i) can directly encode shapes represented as such n-ary graphs, (ii) can be robustly trained on large and complex shape families, and (iii) be used to generate a great diversity of realistic structured shape geometries. Technically, we accomplish this by drawing inspiration from recent advances in graph neural networks to propose an order-invariant encoding of n-ary graphs, considering jointly both part geometry and inter-part relations during network training. We extensively evaluate the quality of the learned latent spaces for various shape families and show significant advantages over baseline and competing methods. The learned latent spaces enable several structure-aware geometry processing applications, including shape generation and interpolation, shape editing, or shape structure discovery directly from un-annotated images, point clouds, or partial scans.

PlanIT: planning and instantiating indoor scenes with relation graph and spatial prior networks

Abstract: We present a new framework for interior scene synthesis that combines a high-level relation graph representation with spatial prior neural networks. We observe that prior work on scene synthesis is divided into two camps: object-oriented approaches (which reason about the set of objects in a scene and their configurations) and space-oriented approaches (which reason about what objects occupy what regions of space). Our insight is that the object-oriented paradigm excels at high-level planning of how a room should be laid out, while the space-oriented paradigm performs well at instantiating a layout by placing objects in precise spatial configurations. With this in mind, we present PlanIT, a layout-generation framework that divides the problem into two distinct planning and instantiation phases. PlanIT represents the "plan" for a scene via a relation graph, encoding objects as nodes and spatial/semantic relationships between objects as edges. In the planning phase, it uses a deep graph convolutional generative model to synthesize relation graphs. In the instantiation phase, it uses image-based convolutional network modules to guide a search procedure that places objects into the scene in a manner consistent with the graph. By decomposing the problem in this way, PlanIT generates scenes of comparable quality to those generated by prior approaches (as judged by both people and learned classifiers), while also providing the modeling flexibility of the intermediate relationship graph representation. These graphs allow the system to support applications such as scene synthesis from a partial graph provided by a user.

Text-based editing of talking-head video

Abstract: Editing talking-head video to change the speech content or to remove filler words is challenging. We propose a novel method to edit talking-head video based on its transcript to produce a realistic output video in which the dialogue of the speaker has been modified, while maintaining a seamless audio-visual flow (i.e. no jump cuts). Our method automatically annotates an input talking-head video with phonemes, visemes, 3D face pose and geometry, reflectance, expression and scene illumination per frame. To edit a video, the user has to only edit the transcript, and an optimization strategy then chooses segments of the input corpus as base material. The annotated parameters corresponding to the selected segments are seamlessly stitched together and used to produce an intermediate video representation in which the lower half of the face is rendered with a parametric face model. Finally, a recurrent video generation network transforms this representation to a photorealistic video that matches the edited transcript. We demonstrate a large variety of edits, such as the addition, removal, and alteration of words, as well as convincing language translation and full sentence synthesis.

DReCon: data-driven responsive control of physics-based characters

Abstract: Interactive control of self-balancing, physically simulated humanoids is a long standing problem in the field of real-time character animation. While physical simulation guarantees realistic interactions in the virtual world, simulated characters can appear unnatural if they perform unusual movements in order to maintain balance. Therefore, obtaining a high level of responsiveness to user control, runtime performance, and diversity has often been overlooked in exchange for motion quality. Recent work in the field of deep reinforcement learning has shown that training physically simulated characters to follow motion capture clips can yield high quality tracking results. We propose a two-step approach for building responsive simulated character controllers from unstructured motion capture data. First, meaningful features from the data such as movement direction, heading direction, speed, and locomotion style, are interactively specified and drive a kinematic character controller implemented using motion matching. Second, reinforcement learning is used to train a simulated character controller that is general enough to track the entire distribution of motion that can be generated by the kinematic controller. Our design emphasizes responsiveness to user input, visual quality, and low runtime cost for application in video-games.

Neural Rendering and Reenactment of Human Actor Videos

Abstract: We propose a method for generating video-realistic animations of real humans under user control. In contrast to conventional human character rendering, we do not require the availability of a production-quality photo-realistic three-dimensional (3D) model of the human but instead rely on a video sequence in conjunction with a (medium-quality) controllable 3D template model of the person. With that, our approach significantly reduces production cost compared to conventional rendering approaches based on production-quality 3D models and can also be used to realistically edit existing videos. Technically, this is achieved by training a neural network that translates simple synthetic images of a human character into realistic imagery. For training our networks, we first track the 3D motion of the person in the video using the template model and subsequently generate a synthetically rendered version of the video. These images are then used to train a conditional generative adversarial network that translates synthetic images of the 3D model into realistic imagery of the human. We evaluate our method for the reenactment of another person that is tracked to obtain the motion data, and show video results generated from artist-designed skeleton motion. Our results outperform the state of the art in learning-based human image synthesis.

Deep inverse rendering for high-resolution SVBRDF estimation from an arbitrary number of images

Abstract: In this paper we present a unified deep inverse rendering framework for estimating the spatially-varying appearance properties of a planar exemplar from an arbitrary number of input photographs, ranging from just a single photograph to many photographs. The precision of the estimated appearance scales from plausible when the input photographs fails to capture all the reflectance information, to accurate for large input sets. A key distinguishing feature of our framework is that it directly optimizes for the appearance parameters in a latent embedded space of spatially-varying appearance, such that no handcrafted heuristics are needed to regularize the optimization. This latent embedding is learned through a fully convolutional auto-encoder that has been designed to regularize the optimization. Our framework not only supports an arbitrary number of input photographs, but also at high resolution. We demonstrate and evaluate our deep inverse rendering solution on a wide variety of publicly available datasets.

Wave-based non-line-of-sight imaging using fast f-k migration

Abstract: Imaging objects outside a camera's direct line of sight has important applications in robotic vision, remote sensing, and many other domains. Time-of-flight-based non-line-of-sight (NLOS) imaging systems have recently demonstrated impressive results, but several challenges remain. Image formation and inversion models have been slow or limited by the types of hidden surfaces that can be imaged. Moreover, non-planar sampling surfaces and non-confocal scanning methods have not been supported by efficient NLOS algorithms. With this work, we introduce a wave-based image formation model for the problem of NLOS imaging. Inspired by inverse methods used in seismology, we adapt a frequency-domain method, f-k migration, for solving the inverse NLOS problem. Unlike existing NLOS algorithms, f-k migration is both fast and memory efficient, it is robust to specular and other complex reflectance properties, and we show how it can be used with non-confocally scanned measurements as well as for non-planar sampling surfaces. f-k migration is more robust to measurement noise than alternative methods, generally produces better quality reconstructions, and is easy to implement. We experimentally validate our algorithms with a new NLOS imaging system that records room-sized scenes outdoors under indirect sunlight, and scans persons wearing retroreflective clothing at interactive rates.

Learning predict-and-simulate policies from unorganized human motion data

Abstract: The goal of this research is to create physically simulated biped characters equipped with a rich repertoire of motor skills. The user can control the characters interactively by modulating their control objectives. The characters can interact physically with each other and with the environment. We present a novel network-based algorithm that learns control policies from unorganized, minimally-labeled human motion data. The network architecture for interactive character animation incorporates an RNN-based motion generator into a DRL-based controller for physics simulation and control. The motion generator guides forward dynamics simulation by feeding a sequence of future motion frames to track. The rich future prediction facilitates policy learning from large training data sets. We will demonstrate the effectiveness of our approach with biped characters that learn a variety of dynamic motor skills from large, unorganized data and react to unexpected perturbation beyond the scope of the training data.

Real-time pose and shape reconstruction of two interacting hands with a single depth camera

Abstract: We present a novel method for real-time pose and shape reconstruction of two strongly interacting hands. Our approach is the first two-hand tracking solution that combines an extensive list of favorable properties, namely it is marker-less, uses a single consumer-level depth camera, runs in real time, handles inter- and intra-hand collisions, and automatically adjusts to the user's hand shape. In order to achieve this, we embed a recent parametric hand pose and shape model and a dense correspondence predictor based on a deep neural network into a suitable energy minimization framework. For training the correspondence prediction network, we synthesize a two-hand dataset based on physical simulations that includes both hand pose and shape annotations while at the same time avoiding inter-hand penetrations. To achieve real-time rates, we phrase the model fitting in terms of a nonlinear least-squares problem so that the energy can be optimized based on a highly efficient GPU-based Gauss-Newton optimizer. We show state-of-the-art results in scenes that exceed the complexity level demonstrated by previous work, including tight two-hand grasps, significant inter-hand occlusions, and gesture interaction.1

The relightables: volumetric performance capture of humans with realistic relighting

Abstract: We present "The Relightables", a volumetric capture system for photorealistic and high quality relightable full-body performance capture. While significant progress has been made on volumetric capture systems, focusing on 3D geometric reconstruction with high resolution textures, much less work has been done to recover photometric properties needed for relighting. Results from such systems lack high-frequency details and the subject's shading is prebaked into the texture. In contrast, a large body of work has addressed relightable acquisition for image-based approaches, which photograph the subject under a set of basis lighting conditions and recombine the images to show the subject as they would appear in a target lighting environment. However, to date, these approaches have not been adapted for use in the context of a high-resolution volumetric capture system. Our method combines this ability to realistically relight humans for arbitrary environments, with the benefits of free-viewpoint volumetric capture and new levels of geometric accuracy for dynamic performances. Our subjects are recorded inside a custom geodesic sphere outfitted with 331 custom color LED lights, an array of high-resolution cameras, and a set of custom high-resolution depth sensors. Our system innovates in multiple areas: First, we designed a novel active depth sensor to capture 12.4 MP depth maps, which we describe in detail. Second, we show how to design a hybrid geometric and machine learning reconstruction pipeline to process the high resolution input and output a volumetric video. Third, we generate temporally consistent reflectance maps for dynamic performers by leveraging the information contained in two alternating color gradient illumination images acquired at 60Hz. Multiple experiments, comparisons, and applications show that The Relightables significantly improves upon the level of realism in placing volumetrically captured human performances into arbitrary CG scenes.

Taichi: a language for high-performance computation on spatially sparse data structures

Abstract: 3D visual computing data are often spatially sparse. To exploit such sparsity, people have developed hierarchical sparse data structures, such as multi-level sparse voxel grids, particles, and 3D hash tables. However, developing and using these high-performance sparse data structures is challenging, due to their intrinsic complexity and overhead. We propose Taichi, a new data-oriented programming language for efficiently authoring, accessing, and maintaining such data structures. The language offers a high-level, data structure-agnostic interface for writing computation code. The user independently specifies the data structure. We provide several elementary components with different sparsity properties that can be arbitrarily composed to create a wide range of multi-level sparse data structures. This decoupling of data structures from computation makes it easy to experiment with different data structures without changing computation code, and allows users to write computation as if they are working with a dense array. Our compiler then uses the semantics of the data structure and index analysis to automatically optimize for locality, remove redundant operations for coherent accesses, maintain sparsity and memory allocations, and generate efficient parallel and vectorized instructions for CPUs and GPUs. Our approach yields competitive performance on common computational kernels such as stencil applications, neighbor lookups, and particle scattering. We demonstrate our language by implementing simulation, rendering, and vision tasks including a material point method simulation, finite element analysis, a multigrid Poisson solver for pressure projection, volumetric path tracing, and 3D convolution on sparse grids. Our computation-data structure decoupling allows us to quickly experiment with different data arrangements, and to develop high-performance data structures tailored for specific computational tasks. With 110 th as many lines of code, we achieve 4.55× higher performance on average, compared to hand-optimized reference implementations.

Data-driven interior plan generation for residential buildings

Abstract: We propose a novel data-driven technique for automatically and efficiently generating floor plans for residential buildings with given boundaries. Central to this method is a two-stage approach that imitates the human design process by locating rooms first and then walls while adapting to the input building boundary. Based on observations of the presence of the living room in almost all floor plans, our designed learning network begins with positioning a living room and continues by iteratively generating other rooms. Then, walls are first determined by an encoder-decoder network, and then they are refined to vector representations using dedicated rules. To effectively train our networks, we construct RPLAN - a manually collected large-scale densely annotated dataset of floor plans from real residential buildings. Intensive experiments, including formative user studies and comparisons, are conducted to illustrate the feasibility and efficacy of our proposed approach. By comparing the plausibility of different floor plans, we have observed that our method substantially outperforms existing methods, and in many cases our floor plans are comparable to human-created ones.

Content-aware generative modeling of graphic design layouts

Abstract: Layout is fundamental to graphic designs. For visual attractiveness and efficient communication of messages and ideas, graphic design layouts often have great variation, driven by the contents to be presented. In this paper, we study the problem of content-aware graphic design layout generation. We propose a deep generative model for graphic design layouts that is able to synthesize layout designs based on the visual and textual semantics of user inputs. Unlike previous approaches that are oblivious to the input contents and rely on heuristic criteria, our model captures the effect of visual and textual contents on layouts, and implicitly learns complex layout structure variations from data without the use of any heuristic rules. To train our model, we build a large-scale magazine layout dataset with fine-grained layout annotations and keyword labeling. Experimental results show that our model can synthesize high-quality layouts based on the visual semantics of input images and keyword-based summary of input text. We also demonstrate that our model internally learns powerful features that capture the subtle interaction between contents and layouts, which are useful for layout-aware design retrieval.

Interactive hand pose estimation using a stretch-sensing soft glove

Abstract: We propose a stretch-sensing soft glove to interactively capture hand poses with high accuracy and without requiring an external optical setup. We demonstrate how our device can be fabricated and calibrated at low cost, using simple tools available in most fabrication labs. To reconstruct the pose from the capacitive sensors embedded in the glove, we propose a deep network architecture that exploits the spatial layout of the sensor itself. The network is trained only once, using an inexpensive off-the-shelf hand pose reconstruction system to gather the training data. The per-user calibration is then performed on-the-fly using only the glove. The glove's capabilities are demonstrated in a series of ablative experiments, exploring different models and calibration methods. Comparing against commercial data gloves, we achieve a 35% improvement in reconstruction accuracy.

Deep view synthesis from sparse photometric images

Abstract: The goal of light transport acquisition is to take images from a sparse set of lighting and viewing directions, and combine them to enable arbitrary relighting with changing view. While relighting from sparse images has received significant attention, there has been relatively less progress on view synthesis from a sparse set of "photometric" images---images captured under controlled conditions, lit by a single directional source; we use a spherical gantry to position the camera on a sphere surrounding the object. In this paper, we synthesize novel viewpoints across a wide range of viewing directions (covering a 60° cone) from a sparse set of just six viewing directions. While our approach relates to previous view synthesis and image-based rendering techniques, those methods are usually restricted to much smaller baselines, and are captured under environment illumination. At our baselines, input images have few correspondences and large occlusions; however we benefit from structured photometric images. Our method is based on a deep convolutional network trained to directly synthesize new views from the six input views. This network combines 3D convolutions on a plane sweep volume with a novel per-view per-depth plane attention map prediction network to effectively aggregate multi-view appearance. We train our network with a large-scale synthetic dataset of 1000 scenes with complex geometry and material properties. In practice, it is able to synthesize novel viewpoints for captured real data and reproduces complex appearance effects like occlusions, view-dependent specularities and hard shadows. Moreover, the method can also be combined with previous relighting techniques to enable changing both lighting and view, and applied to computer vision problems like multiview stereo from sparse image sets.