Showing papers in "ACM Transactions on Graphics in 2018"

DeepMimic: example-guided deep reinforcement learning of physics-based character skills

Abstract: A longstanding goal in character animation is to combine data-driven specification of behavior with a system that can execute a similar behavior in a physical simulation, thus enabling realistic responses to perturbations and environmental variation. We show that well-known reinforcement learning (RL) methods can be adapted to learn robust control policies capable of imitating a broad range of example motion clips, while also learning complex recoveries, adapting to changes in morphology, and accomplishing user-specified goals. Our method handles keyframed motions, highly-dynamic actions such as motion-captured flips and spins, and retargeted motions. By combining a motion-imitation objective with a task objective, we can train characters that react intelligently in interactive settings, e.g., by walking in a desired direction or throwing a ball at a user-specified target. This approach thus combines the convenience and motion quality of using motion clips to define the desired style and appearance, with the flexibility and generality afforded by RL methods and physics-based animation. We further explore a number of methods for integrating multiple clips into the learning process to develop multi-skilled agents capable of performing a rich repertoire of diverse skills. We demonstrate results using multiple characters (human, Atlas robot, bipedal dinosaur, dragon) and a large variety of skills, including locomotion, acrobatics, and martial arts.

Looking to listen at the cocktail party: a speaker-independent audio-visual model for speech separation

Abstract: We present a joint audio-visual model for isolating a single speech signal from a mixture of sounds such as other speakers and background noise. Solving this task using only audio as input is extremely challenging and does not provide an association of the separated speech signals with speakers in the video. In this paper, we present a deep network-based model that incorporates both visual and auditory signals to solve this task. The visual features are used to "focus" the audio on desired speakers in a scene and to improve the speech separation quality. To train our joint audio-visual model, we introduce AVSpeech, a new dataset comprised of thousands of hours of video segments from the Web. We demonstrate the applicability of our method to classic speech separation tasks, as well as real-world scenarios involving heated interviews, noisy bars, and screaming children, only requiring the user to specify the face of the person in the video whose speech they want to isolate. Our method shows clear advantage over state-of-the-art audio-only speech separation in cases of mixed speech. In addition, our model, which is speaker-independent (trained once, applicable to any speaker), produces better results than recent audio-visual speech separation methods that are speaker-dependent (require training a separate model for each speaker of interest).

Stereo magnification: learning view synthesis using multiplane images

Abstract: The view synthesis problem---generating novel views of a scene from known imagery---has garnered recent attention due in part to compelling applications in virtual and augmented reality. In this paper, we explore an intriguing scenario for view synthesis: extrapolating views from imagery captured by narrow-baseline stereo cameras, including VR cameras and now-widespread dual-lens camera phones. We call this problem stereo magnification, and propose a learning framework that leverages a new layered representation that we call multiplane images (MPIs). Our method also uses a massive new data source for learning view extrapolation: online videos on YouTube. Using data mined from such videos, we train a deep network that predicts an MPI from an input stereo image pair. This inferred MPI can then be used to synthesize a range of novel views of the scene, including views that extrapolate significantly beyond the input baseline. We show that our method compares favorably with several recent view synthesis methods, and demonstrate applications in magnifying narrow-baseline stereo images.

Point convolutional neural networks by extension operators

Abstract: This paper presents Point Convolutional Neural Networks (PCNN): a novel framework for applying convolutional neural networks to point clouds. The framework consists of two operators: extension and restriction, mapping point cloud functions to volumetric functions and vise-versa. A point cloud convolution is defined by pull-back of the Euclidean volumetric convolution via an extension-restriction mechanism. The point cloud convolution is computationally efficient, invariant to the order of points in the point cloud, robust to different samplings and varying densities, and translation invariant, that is the same convolution kernel is used at all points. PCNN generalizes image CNNs and allows readily adapting their architectures to the point cloud setting. Evaluation of PCNN on three central point cloud learning benchmarks convincingly outperform competing point cloud learning methods, and the vast majority of methods working with more informative shape representations such as surfaces and/or normals.

Differentiable Monte Carlo ray tracing through edge sampling

Abstract: Gradient-based methods are becoming increasingly important for computer graphics, machine learning, and computer vision. The ability to compute gradients is crucial to optimization, inverse problems, and deep learning. In rendering, the gradient is required with respect to variables such as camera parameters, light sources, scene geometry, or material appearance. However, computing the gradient of rendering is challenging because the rendering integral includes visibility terms that are not differentiable. Previous work on differentiable rendering has focused on approximate solutions. They often do not handle secondary effects such as shadows or global illumination, or they do not provide the gradient with respect to variables other than pixel coordinates.We introduce a general-purpose differentiable ray tracer, which, to our knowledge, is the first comprehensive solution that is able to compute derivatives of scalar functions over a rendered image with respect to arbitrary scene parameters such as camera pose, scene geometry, materials, and lighting parameters. The key to our method is a novel edge sampling algorithm that directly samples the Dirac delta functions introduced by the derivatives of the discontinuous integrand. We also develop efficient importance sampling methods based on spatial hierarchies. Our method can generate gradients in times running from seconds to minutes depending on scene complexity and desired precision.We interface our differentiable ray tracer with the deep learning library PyTorch and show prototype applications in inverse rendering and the generation of adversarial examples for neural networks.

End-to-end optimization of optics and image processing for achromatic extended depth of field and super-resolution imaging

Abstract: In typical cameras the optical system is designed first; once it is fixed, the parameters in the image processing algorithm are tuned to get good image reproduction. In contrast to this sequential design approach, we consider joint optimization of an optical system (for example, the physical shape of the lens) together with the parameters of the reconstruction algorithm. We build a fully-differentiable simulation model that maps the true source image to the reconstructed one. The model includes diffractive light propagation, depth and wavelength-dependent effects, noise and nonlinearities, and the image post-processing. We jointly optimize the optical parameters and the image processing algorithm parameters so as to minimize the deviation between the true and reconstructed image, over a large set of images. We implement our joint optimization method using autodifferentiation to efficiently compute parameter gradients in a stochastic optimization algorithm. We demonstrate the efficacy of this approach by applying it to achromatic extended depth of field and snapshot super-resolution imaging.

Deep blending for free-viewpoint image-based rendering

Abstract: Free-viewpoint image-based rendering (IBR) is a standing challenge. IBR methods combine warped versions of input photos to synthesize a novel view. The image quality of this combination is directly affected by geometric inaccuracies of multi-view stereo (MVS) reconstruction and by view- and image-dependent effects that produce artifacts when contributions from different input views are blended. We present a new deep learning approach to blending for IBR, in which we use held-out real image data to learn blending weights to combine input photo contributions. Our Deep Blending method requires us to address several challenges to achieve our goal of interactive free-viewpoint IBR navigation. We first need to provide sufficiently accurate geometry so the Convolutional Neural Network (CNN) can succeed in finding correct blending weights. We do this by combining two different MVS reconstructions with complementary accuracy vs. completeness tradeoffs. To tightly integrate learning in an interactive IBR system, we need to adapt our rendering algorithm to produce a fixed number of input layers that can then be blended by the CNN. We generate training data with a variety of captured scenes, using each input photo as ground truth in a held-out approach. We also design the network architecture and the training loss to provide high quality novel view synthesis, while reducing temporal flickering artifacts. Our results demonstrate free-viewpoint IBR in a wide variety of scenes, clearly surpassing previous methods in visual quality, especially when moving far from the input cameras.

Deep appearance models for face rendering

Abstract: We introduce a deep appearance model for rendering the human face. Inspired by Active Appearance Models, we develop a data-driven rendering pipeline that learns a joint representation of facial geometry and appearance from a multiview capture setup. Vertex positions and view-specific textures are modeled using a deep variational autoencoder that captures complex nonlinear effects while producing a smooth and compact latent representation. View-specific texture enables the modeling of view-dependent effects such as specularity. In addition, it can also correct for imperfect geometry stemming from biased or low resolution estimates. This is a significant departure from the traditional graphics pipeline, which requires highly accurate geometry as well as all elements of the shading model to achieve realism through physically-inspired light transport. Acquiring such a high level of accuracy is difficult in practice, especially for complex and intricate parts of the face, such as eyelashes and the oral cavity. These are handled naturally by our approach, which does not rely on precise estimates of geometry. Instead, the shading model accommodates deficiencies in geometry though the flexibility afforded by the neural network employed. At inference time, we condition the decoding network on the viewpoint of the camera in order to generate the appropriate texture for rendering. The resulting system can be implemented simply using existing rendering engines through dynamic textures with flat lighting. This representation, together with a novel unsupervised technique for mapping images to facial states, results in a system that is naturally suited to real-time interactive settings such as Virtual Reality (VR).

Learning to reconstruct shape and spatially-varying reflectance from a single image

Abstract: Reconstructing shape and reflectance properties from images is a highly under-constrained problem, and has previously been addressed by using specialized hardware to capture calibrated data or by assuming known (or highly constrained) shape or reflectance. In contrast, we demonstrate that we can recover non-Lambertian, spatially-varying BRDFs and complex geometry belonging to any arbitrary shape class, from a single RGB image captured under a combination of unknown environment illumination and flash lighting. We achieve this by training a deep neural network to regress shape and reflectance from the image. Our network is able to address this problem because of three novel contributions: first, we build a large-scale dataset of procedurally generated shapes and real-world complex SVBRDFs that approximate real world appearance well. Second, single image inverse rendering requires reasoning at multiple scales, and we propose a cascade network structure that allows this in a tractable manner. Finally, we incorporate an in-network rendering layer that aids the reconstruction task by handling global illumination effects that are important for real-world scenes. Together, these contributions allow us to tackle the entire inverse rendering problem in a holistic manner and produce state-of-the-art results on both synthetic and real data.

Deep exemplar-based colorization

Abstract: We propose the first deep learning approach for exemplar-based local colorization. Given a reference color image, our convolutional neural network directly maps a grayscale image to an output colorized image. Rather than using hand-crafted rules as in traditional exemplar-based methods, our end-to-end colorization network learns how to select, propagate, and predict colors from the large-scale data. The approach performs robustly and generalizes well even when using reference images that are unrelated to the input grayscale image. More importantly, as opposed to other learning-based colorization methods, our network allows the user to achieve customizable results by simply feeding different references. In order to further reduce manual effort in selecting the references, the system automatically recommends references with our proposed image retrieval algorithm, which considers both semantic and luminance information. The colorization can be performed fully automatically by simply picking the top reference suggestion. Our approach is validated through a user study and favorable quantitative comparisons to the-state-of-the-art methods. Furthermore, our approach can be naturally extended to video colorization. Our code and models are freely available for public use.

Exposure: A White-Box Photo Post-Processing Framework

Abstract: Retouching can significantly elevate the visual appeal of photos, but many casual photographers lack the expertise to do this well. To address this problem, previous works have proposed automatic retouching systems based on supervised learning from paired training images acquired before and after manual editing. As it is difficult for users to acquire paired images that reflect their retouching preferences, we present in this article a deep learning approach that is instead trained on unpaired data, namely, a set of photographs that exhibits a retouching style the user likes, which is much easier to collect. Our system is formulated using deep convolutional neural networks that learn to apply different retouching operations on an input image. Network training with respect to various types of edits is enabled by modeling these retouching operations in a unified manner as resolution-independent differentiable filters. To apply the filters in a proper sequence and with suitable parameters, we employ a deep reinforcement learning approach that learns to make decisions on what action to take next, given the current state of the image. In contrast to many deep learning systems, ours provides users with an understandable solution in the form of conventional retouching edits rather than just a “black-box” result. Through quantitative comparisons and user studies, we show that this technique generates retouching results consistent with the provided photo set.

Mode-adaptive neural networks for quadruped motion control

Abstract: Quadruped motion includes a wide variation of gaits such as walk, pace, trot and canter, and actions such as jumping, sitting, turning and idling. Applying existing data-driven character control frameworks to such data requires a significant amount of data preprocessing such as motion labeling and alignment. In this paper, we propose a novel neural network architecture called Mode-Adaptive Neural Networks for controlling quadruped characters. The system is composed of the motion prediction network and the gating network. At each frame, the motion prediction network computes the character state in the current frame given the state in the previous frame and the user-provided control signals. The gating network dynamically updates the weights of the motion prediction network by selecting and blending what we call the expert weights, each of which specializes in a particular movement. Due to the increased flexibility, the system can learn consistent expert weights across a wide range of non-periodic/periodic actions, from unstructured motion capture data, in an end-to-end fashion. In addition, the users are released from performing complex labeling of phases in different gaits. We show that this architecture is suitable for encoding the multi-modality of quadruped locomotion and synthesizing responsive motion in real-time.

Monte Carlo convolution for learning on non-uniformly sampled point clouds

Abstract: Deep learning systems extensively use convolution operations to process input data. Though convolution is clearly defined for structured data such as 2D images or 3D volumes, this is not true for other data types such as sparse point clouds. Previous techniques have developed approximations to convolutions for restricted conditions. Unfortunately, their applicability is limited and cannot be used for general point clouds. We propose an efficient and effective method to learn convolutions for non-uniformly sampled point clouds, as they are obtained with modern acquisition techniques. Learning is enabled by four key novelties: first, representing the convolution kernel itself as a multilayer perceptron; second, phrasing convolution as a Monte Carlo integration problem, third, using this notion to combine information from multiple samplings at different levels; and fourth using Poisson disk sampling as a scalable means of hierarchical point cloud learning. The key idea across all these contributions is to guarantee adequate consideration of the underlying non-uniform sample distribution function from a Monte Carlo perspective. To make the proposed concepts applicable to real-world tasks, we furthermore propose an efficient implementation which significantly reduces the GPU memory required during the training process. By employing our method in hierarchical network architectures we can outperform most of the state-of-the-art networks on established point cloud segmentation, classification and normal estimation benchmarks. Furthermore, in contrast to most existing approaches, we also demonstrate the robustness of our method with respect to sampling variations, even when training with uniformly sampled data only. To support the direct application of these concepts, we provide a ready-to-use TensorFlow implementation of these layers at https://github.com/viscom-ulm/MCCNN.

tempoGAN: a temporally coherent, volumetric GAN for super-resolution fluid flow

Abstract: We propose a temporally coherent generative model addressing the super-resolution problem for fluid flows. Our work represents a first approach to synthesize four-dimensional physics fields with neural networks. Based on a conditional generative adversarial network that is designed for the inference of three-dimensional volumetric data, our model generates consistent and detailed results by using a novel temporal discriminator, in addition to the commonly used spatial one. Our experiments show that the generator is able to infer more realistic high-resolution details by using additional physical quantities, such as low-resolution velocities or vorticities. Besides improvements in the training process and in the generated outputs, these inputs offer means for artistic control as well. We additionally employ a physics-aware data augmentation step, which is crucial to avoid overfitting and to reduce memory requirements. In this way, our network learns to generate adverted quantities with highly detailed, realistic, and temporally coherent features. Our method works instantaneously, using only a single time-step of low-resolution fluid data. We demonstrate the abilities of our method using a variety of complex inputs and applications in two and three dimensions.

Tetrahedral meshing in the wild

Abstract: We propose a novel tetrahedral meshing technique that is unconditionally robust, requires no user interaction, and can directly convert a triangle soup into an analysis-ready volumetric mesh. The approach is based on several core principles: (1) initial mesh construction based on a fully robust, yet efficient, filtered exact computation (2) explicit (automatic or user-defined) tolerancing of the mesh relative to the surface input (3) iterative mesh improvement with guarantees, at every step, of the output validity. The quality of the resulting mesh is a direct function of the target mesh size and allowed tolerance: increasing allowed deviation from the initial mesh and decreasing the target edge length both lead to higher mesh quality. Our approach enables "black-box" analysis, i.e. it allows to automatically solve partial differential equations on geometrical models available in the wild, offering a robustness and reliability comparable to, e.g., image processing algorithms, opening the door to automatic, large scale processing of real-world geometric data.

Single-image SVBRDF capture with a rendering-aware deep network

Abstract: Texture, highlights, and shading are some of many visual cues that allow humans to perceive material appearance in single pictures. Yet, recovering spatially-varying bi-directional reflectance distribution functions (SVBRDFs) from a single image based on such cues has challenged researchers in computer graphics for decades. We tackle lightweight appearance capture by training a deep neural network to automatically extract and make sense of these visual cues. Once trained, our network is capable of recovering per-pixel normal, diffuse albedo, specular albedo and specular roughness from a single picture of a flat surface lit by a hand-held flash. We achieve this goal by introducing several innovations on training data acquisition and network design. For training, we leverage a large dataset of artist-created, procedural SVBRDFs which we sample and render under multiple lighting directions. We further amplify the data by material mixing to cover a wide diversity of shading effects, which allows our network to work across many material classes. Motivated by the observation that distant regions of a material sample often offer complementary visual cues, we design a network that combines an encoder-decoder convolutional track for local feature extraction with a fully-connected track for global feature extraction and propagation. Many important material effects are view-dependent, and as such ambiguous when observed in a single image. We tackle this challenge by defining the loss as a differentiable SVBRDF similarity metric that compares the renderings of the predicted maps against renderings of the ground truth from several lighting and viewing directions. Combined together, these novel ingredients bring clear improvement over state of the art methods for single-shot capture of spatially varying BRDFs.

paGAN: real-time avatars using dynamic textures

Abstract: With the rising interest in personalized VR and gaming experiences comes the need to create high quality 3D avatars that are both low-cost and variegated. Due to this, building dynamic avatars from a single unconstrained input image is becoming a popular application. While previous techniques that attempt this require multiple input images or rely on transferring dynamic facial appearance from a source actor, we are able to do so using only one 2D input image without any form of transfer from a source image. We achieve this using a new conditional Generative Adversarial Network design that allows fine-scale manipulation of any facial input image into a new expression while preserving its identity. Our photoreal avatar GAN (paGAN) can also synthesize the unseen mouth interior and control the eye-gaze direction of the output, as well as produce the final image from a novel viewpoint. The method is even capable of generating fully-controllable temporally stable video sequences, despite not using temporal information during training. After training, we can use our network to produce dynamic image-based avatars that are controllable on mobile devices in real time. To do this, we compute a fixed set of output images that correspond to key blendshapes, from which we extract textures in UV space. Using a subject's expression blendshapes at run-time, we can linearly blend these key textures together to achieve the desired appearance. Furthermore, we can use the mouth interior and eye textures produced by our network to synthesize on-the-fly avatar animations for those regions. Our work produces state-of-the-art quality image and video synthesis, and is the first to our knowledge that is able to generate a dynamically textured avatar with a mouth interior, all from a single image.

Non-stationary texture synthesis by adversarial expansion

Abstract: The real world exhibits an abundance of non-stationary textures. Examples include textures with large scale structures, as well as spatially variant and inhomogeneous textures. While existing example-based texture synthesis methods can cope well with stationary textures, non-stationary textures still pose a considerable challenge, which remains unresolved. In this paper, we propose a new approach for example-based non-stationary texture synthesis. Our approach uses a generative adversarial network (GAN), trained to double the spatial extent of texture blocks extracted from a specific texture exemplar. Once trained, the fully convolutional generator is able to expand the size of the entire exemplar, as well as of any of its sub-blocks. We demonstrate that this conceptually simple approach is highly effective for capturing large scale structures, as well as other non-stationary attributes of the input exemplar. As a result, it can cope with challenging textures, which, to our knowledge, no other existing method can handle.

Deep convolutional priors for indoor scene synthesis

Abstract: We present a convolutional neural network based approach for indoor scene synthesis. By representing 3D scenes with a semantically-enriched image-based representation based on orthographic top-down views, we learn convolutional object placement priors from the entire context of a room. Our approach iteratively generates rooms from scratch, given only the room architecture as input. Through a series of perceptual studies we compare the plausibility of scenes generated using our method against baselines for object selection and object arrangement, as well as scenes modeled by people. We find that our method generates scenes that are preferred over the baselines, and in some cases are equally preferred to human-created scenes.

SFV: reinforcement learning of physical skills from videos

Abstract: Data-driven character animation based on motion capture can produce highly naturalistic behaviors and, when combined with physics simulation, can provide for natural procedural responses to physical perturbations, environmental changes, and morphological discrepancies. Motion capture remains the most popular source of motion data, but collecting mocap data typically requires heavily instrumented environments and actors. In this paper, we propose a method that enables physically simulated characters to learn skills from videos (SFV). Our approach, based on deep pose estimation and deep reinforcement learning, allows data-driven animation to leverage the abundance of publicly available video clips from the web, such as those from YouTube. This has the potential to enable fast and easy design of character controllers simply by querying for video recordings of the desired behavior. The resulting controllers are robust to perturbations, can be adapted to new settings, can perform basic object interactions, and can be retargeted to new morphologies via reinforcement learning. We further demonstrate that our method can predict potential human motions from still images, by forward simulation of learned controllers initialized from the observed pose. Our framework is able to learn a broad range of dynamic skills, including locomotion, acrobatics, and martial arts. (Video1)

Support-free volume printing by multi-axis motion

Abstract: This paper presents a new method to fabricate 3D models on a robotic printing system equipped with multi-axis motion. Materials are accumulated inside the volume along curved tool-paths so that the need of supporting structures can be tremendously reduced - if not completely abandoned - on all models. Our strategy to tackle the challenge of tool-path planning for multi-axis 3D printing is to perform two successive decompositions, first volume-to-surfaces and then surfaces-to-curves. The volume-to-surfaces decomposition is achieved by optimizing a scalar field within the volume that represents the fabrication sequence. The field is constrained such that its iso-values represent curved layers that are supported from below, and present a convex surface affording for collision-free navigation of the printer head. After extracting all curved layers, the surfaces-to-curves decomposition covers them with tool-paths while taking into account constraints from the robotic printing system. Our method successfully generates tool-paths for 3D printing models with large overhangs and high-genus topology. We fabricated several challenging cases on our robotic platform to verify and demonstrate its capabilities.

A moving least squares material point method with displacement discontinuity and two-way rigid body coupling

Abstract: In this paper, we introduce the Moving Least Squares Material Point Method (MLS-MPM). MLS-MPM naturally leads to the formulation of Affine Particle-In-Cell (APIC) [Jiang et al. 2015] and Polynomial Particle-In-Cell [Fu et al. 2017] in a way that is consistent with a Galerkin-style weak form discretization of the governing equations. Additionally, it enables a new stress divergence discretization that effortlessly allows all MPM simulations to run two times faster than before. We also develop a Compatible Particle-In-Cell (CPIC) algorithm on top of MLS-MPM. Utilizing a colored distance field representation and a novel compatibility condition for particles and grid nodes, our framework enables the simulation of various new phenomena that are not previously supported by MPM, including material cutting, dynamic open boundaries, and two-way coupling with rigid bodies. MLS-MPM with CPIC is easy to implement and friendly to performance optimization.

Deep image-based relighting from optimal sparse samples

Abstract: We present an image-based relighting method that can synthesize scene appearance under novel, distant illumination from the visible hemisphere, from only five images captured under pre-defined directional lights. Our method uses a deep convolutional neural network to regress the relit image from these five images; this relighting network is trained on a large synthetic dataset comprised of procedurally generated shapes with real-world reflectances. We show that by combining a custom-designed sampling network with the relighting network, we can jointly learn both the optimal input light directions and the relighting function. We present an extensive evaluation of our network, including an empirical analysis of reconstruction quality, optimal lighting configurations for different scenarios, and alternative network architectures. We demonstrate, on both synthetic and real scenes, that our method is able to reproduce complex, high-frequency lighting effects like specularities and cast shadows, and outperforms other image-based relighting methods that require an order of magnitude more images.

Learning symmetric and low-energy locomotion

Abstract: Learning locomotion skills is a challenging problem. To generate realistic and smooth locomotion, existing methods use motion capture, finite state machines or morphology-specific knowledge to guide the motion generation algorithms. Deep reinforcement learning (DRL) is a promising approach for the automatic creation of locomotion control. Indeed, a standard benchmark for DRL is to automatically create a running controller for a biped character from a simple reward function [Duan et al. 2016]. Although several different DRL algorithms can successfully create a running controller, the resulting motions usually look nothing like a real runner. This paper takes a minimalist learning approach to the locomotion problem, without the use of motion examples, finite state machines, or morphology-specific knowledge. We introduce two modifications to the DRL approach that, when used together, produce locomotion behaviors that are symmetric, low-energy, and much closer to that of a real person. First, we introduce a new term to the loss function (not the reward function) that encourages symmetric actions. Second, we introduce a new curriculum learning method that provides modulated physical assistance to help the character with left/right balance and forward movement. The algorithm automatically computes appropriate assistance to the character and gradually relaxes this assistance, so that eventually the character learns to move entirely without help. Because our method does not make use of motion capture data, it can be applied to a variety of character morphologies. We demonstrate locomotion controllers for the lower half of a biped, a full humanoid, a quadruped, and a hexapod. Our results show that learned policies are able to produce symmetric, low-energy gaits. In addition, speed-appropriate gait patterns emerge without any guidance from motion examples or contact planning.

Denoising with kernel prediction and asymmetric loss functions

Abstract: We present a modular convolutional architecture for denoising rendered images. We expand on the capabilities of kernel-predicting networks by combining them with a number of task-specific modules, and optimizing the assembly using an asymmetric loss. The source-aware encoder---the first module in the assembly---extracts low-level features and embeds them into a common feature space, enabling quick adaptation of a trained network to novel data. The spatial and temporal modules extract abstract, high-level features for kernel-based reconstruction, which is performed at three different spatial scales to reduce low-frequency artifacts. The complete network is trained using a class of asymmetric loss functions that are designed to preserve details and provide the user with a direct control over the variance-bias trade-off during inference. We also propose an error-predicting module for inferring reconstruction error maps that can be used to drive adaptive sampling. Finally, we present a theoretical analysis of convergence rates of kernel-predicting architectures, shedding light on why kernel prediction performs better than synthesizing the colors directly, complementing the empirical evidence presented in this and previous works. We demonstrate that our networks attain results that compare favorably to state-of-the-art methods in terms of detail preservation, low-frequency noise removal, and temporal stability on a variety of production and academic datasets.

Towards virtual reality infinite walking: dynamic saccadic redirection

Abstract: Redirected walking techniques can enhance the immersion and visual-vestibular comfort of virtual reality (VR) navigation, but are often limited by the size, shape, and content of the physical environments. We propose a redirected walking technique that can apply to small physical environments with static or dynamic obstacles. Via a head- and eye-tracking VR headset, our method detects saccadic suppression and redirects the users during the resulting temporary blindness. Our dynamic path planning runs in real-time on a GPU, and thus can avoid static and dynamic obstacles, including walls, furniture, and other VR users sharing the same physical space. To further enhance saccadic redirection, we propose subtle gaze direction methods tailored for VR perception. We demonstrate that saccades can significantly increase the rotation gains during redirection without introducing visual distortions or simulator sickness. This allows our method to apply to large open virtual spaces and small physical environments for room-scale VR. We evaluate our system via numerical simulations and real user studies.

Single-photon 3D imaging with deep sensor fusion

Abstract: Sensors which capture 3D scene information provide useful data for tasks in vehicle navigation, gesture recognition, human pose estimation, and geometric reconstruction. Active illumination time-of-flight sensors in particular have become widely used to estimate a 3D representation of a scene. However, the maximum range, density of acquired spatial samples, and overall acquisition time of these sensors is fundamentally limited by the minimum signal required to estimate depth reliably. In this paper, we propose a data-driven method for photon-efficient 3D imaging which leverages sensor fusion and computational reconstruction to rapidly and robustly estimate a dense depth map from low photon counts. Our sensor fusion approach uses measurements of single photon arrival times from a low-resolution single-photon detector array and an intensity image from a conventional high-resolution camera. Using a multi-scale deep convolutional network, we jointly process the raw measurements from both sensors and output a high-resolution depth map. To demonstrate the efficacy of our approach, we implement a hardware prototype and show results using captured data. At low signal-to-background levels, our depth reconstruction algorithm with sensor fusion outperforms other methods for depth estimation from noisy measurements of photon arrival times.

Faceshop: deep sketch-based face image editing

Abstract: We present a novel system for sketch-based face image editing, enabling users to edit images intuitively by sketching a few strokes on a region of interest. Our interface features tools to express a desired image manipulation by providing both geometry and color constraints as user-drawn strokes. As an alternative to the direct user input, our proposed system naturally supports a copy-paste mode, which allows users to edit a given image region by using parts of another exemplar image without the need of hand-drawn sketching at all. The proposed interface runs in real-time and facilitates an interactive and iterative workflow to quickly express the intended edits. Our system is based on a novel sketch domain and a convolutional neural network trained end-to-end to automatically learn to render image regions corresponding to the input strokes. To achieve high quality and semantically consistent results we train our neural network on two simultaneous tasks, namely image completion and image translation. To the best of our knowledge, we are the first to combine these two tasks in a unified framework for interactive image editing. Our results show that the proposed sketch domain, network architecture, and training procedure generalize well to real user input and enable high quality synthesis results without additional post-processing.

Deep inertial poser: learning to reconstruct human pose from sparse inertial measurements in real time

Abstract: We demonstrate a novel deep neural network capable of reconstructing human full body pose in real-time from 6 Inertial Measurement Units (IMUs) worn on the user's body. In doing so, we address several difficult challenges. First, the problem is severely under-constrained as multiple pose parameters produce the same IMU orientations. Second, capturing IMU data in conjunction with ground-truth poses is expensive and difficult to do in many target application scenarios (e.g., outdoors). Third, modeling temporal dependencies through non-linear optimization has proven effective in prior work but makes real-time prediction infeasible. To address this important limitation, we learn the temporal pose priors using deep learning. To learn from sufficient data, we synthesize IMU data from motion capture datasets. A bi-directional RNN architecture leverages past and future information that is available at training time. At test time, we deploy the network in a sliding window fashion, retaining real time capabilities. To evaluate our method, we recorded DIP-IMU, a dataset consisting of 10 subjects wearing 17 IMUs for validation in 64 sequences with 330 000 time instants; this constitutes the largest IMU dataset publicly available. We quantitatively evaluate our approach on multiple datasets and show results from a real-time implementation. DIP-IMU and the code are available for research purposes.1

High-fidelity facial reflectance and geometry inference from an unconstrained image

Abstract: We present a deep learning-based technique to infer high-quality facial reflectance and geometry given a single unconstrained image of the subject, which may contain partial occlusions and arbitrary illumination conditions. The reconstructed high-resolution textures, which are generated in only a few seconds, include high-resolution skin surface reflectance maps, representing both the diffuse and specular albedo, and medium- and high-frequency displacement maps, thereby allowing us to render compelling digital avatars under novel lighting conditions. To extract this data, we train our deep neural networks with a high-quality skin reflectance and geometry database created with a state-of-the-art multi-view photometric stereo system using polarized gradient illumination. Given the raw facial texture map extracted from the input image, our neural networks synthesize complete reflectance and displacement maps, as well as complete missing regions caused by occlusions. The completed textures exhibit consistent quality throughout the face due to our network architecture, which propagates texture features from the visible region, resulting in high-fidelity details that are consistent with those seen in visible regions. We describe how this highly underconstrained problem is made tractable by dividing the full inference into smaller tasks, which are addressed by dedicated neural networks. We demonstrate the effectiveness of our network design with robust texture completion from images of faces that are largely occluded. With the inferred reflectance and geometry data, we demonstrate the rendering of high-fidelity 3D avatars from a variety of subjects captured under different lighting conditions. In addition, we perform evaluations demonstrating that our method can infer plausible facial reflectance and geometric details comparable to those obtained from high-end capture devices, and outperform alternative approaches that require only a single unconstrained input image.