Search

3D-aware portrait lighting control is an emerging and promising domain, thanks to the recent advance of generative adversarial networks and neural radiance fields. Existing solutions typically try to decouple the lighting from the geometry and appearance for disentangled control with an explicit lighting representation (e.g., Lambertian or Phong). However, they either are limited to a constrained lighting condition (e.g., directional light) or demand a tricky-to-fetch dataset as supervision for the intrinsic compositions (e.g., the albedo). We propose NeRFFaceLighting to explore an implicit representation for portrait lighting based on the pretrained tri-plane representation to address the above limitations. We approach this disentangled lighting-control problem by distilling the shading from the original fused representation of both appearance and lighting (i.e., one tri-plane) to their disentangled representations (i.e., two tri-planes) with the conditional discriminator to supervise the lighting effects. We further carefully design the regularization to reduce the ambiguity of such decomposition and enhance the ability of generalization to unseen lighting conditions. Moreover, our method can be extended to enable 3D-aware real portrait relighting. Through extensive quantitative and qualitative evaluations, we demonstrate the superior 3D-aware lighting control ability of our model compared to alternative and existing solutions.

Realistic 3D facial generation based on Neural Radiance Fields (NeRFs) from 2D sketches benefits various applications. Despite the high realism of free-view rendering results of NeRFs, it is tedious and difficult for artists to achieve detailed 3D control and manipulation. Meanwhile, due to its conciseness and expressiveness, sketching has been widely used for 2D facial image generation and editing. Applying sketching to NeRFs is challenging due to the inherent uncertainty for 3D generation with 2D constraints, a significant gap in content richness when generating faces from sparse sketches, and potential inconsistencies for sequential multi-view editing given only 2D sketch inputs. To address these challenges, we present SketchFaceNeRF, a novel sketch-based 3D facial NeRF generation and editing method, to produce free-view photo-realistic images. To solve the challenge of sketch sparsity, we introduce a Sketch Tri-plane Prediction net to first inject the appearance into sketches, thus generating features given reference images to allow color and texture control. Such features are then lifted into compact 3D tri-planes to supplement the absent 3D information, which is important for improving robustness and faithfulness. However, during editing, consistency for unseen or unedited 3D regions is difficult to maintain due to limited spatial hints in sketches. We thus adopt a Mask Fusion module to transform free-view 2D masks (inferred from sketch editing operations) into the tri-plane space as 3D masks, which guide the fusion of the original and sketch-based generated faces to synthesize edited faces. We further design an optimization approach with a novel space loss to improve identity retention and editing faithfulness. Our pipeline enables users to flexibly manipulate faces from different viewpoints in 3D space, easily designing desirable facial models. Extensive experiments validate that our approach is superior to the state-of-The-Art 2D sketch-based image generati

Text input is a desired feature for AR glasses. While there already exist various input modalities (e.g., voice, mid-air gesture), the diverse demands required by different input scenarios can hardly be met by the small number of fixed input postures offered by existing solutions. In this paper, we present Handwriting Velcro, a novel text input solution for AR glasses based on flexible touch sensors. The distinct advantage of our system is that it can easily stick to different body parts, thus endowing AR glasses with posture-adaptive handwriting input. We explored the design space of on-body device positions and identified the best interaction positions for various user postures. To flatten users' learning curves, we adapt our device to the established writing habits of different users by training a 36-character (i.e., A-Z, 0-9) recognition neural network in a human-in-the-loop manner. Such a personalization attempt ultimately achieves a low error rate of 0.005 on average for users with different writing styles. Subjective feedback shows that our solution has a good performance in system practicability and social acceptance. Empirically, we conducted a heuristic study to explore and identify the best interaction Position-Posture Correlation. Experimental results show that our Handwriting Velcro excels similar work [6] and commercial product in both practicality (12.3 WPM) and user-friendliness in different contexts. © 2023 ACM.

We propose a new method for realistic human motion transfer using a generative adversarial network (GAN), which generates a motion video of a target character imitating actions of a source character, while maintaining high authenticity of the generated results. We tackle the problem by decoupling and recombining the posture information and appearance information of both the source and target characters. The innovation of our approach lies in the use of the projection of a reconstructed 3D human model as the condition of GAN to better maintain the structural integrity of transfer results in different poses. We further introduce a detail enhancement net to enhance the details of transfer results by exploiting the details in real source frames. Extensive experiments show that our approach yields better results both qualitatively and quantitatively than the state-of-the-art methods.

Commercial scenes such as markets and shops are everyday scenes for both virtual scenes and real-world interior designs. However, existing literature on interior scene synthesis mainly focuses on formulating and optimizing residential scenes such as bedrooms, living rooms, etc. Existing literature typically presents a set of relations among objects. It recognizes each furniture object as the smallest unit while optimizing a residential room. However, object relations become less critical in commercial scenes since shelves are often placed next to each other so pre-calculated relations of objects are less needed. Instead, interior designers resort to evaluating how groups of objects perform in commercial scenes, i.e., the smallest unit to be evaluated is a group of objects. This paper presents a system automatically synthesizes market-like commercial scenes in virtual environments. Following the rules of commercial layout design, we parameterize groups of objects as "patterns"contributing to a scene. Each pattern directly yields a human-centric routine locally, provides potential connectivity with other routines, and derives the arrangements of objects concerning itself according to the assigned parameters. In order to optimize a scene, the patterns are iteratively multiplexed to insert new routines or modify existing ones under a set of constraints derived from commercial layout designs. Through extensive experiments, we demonstrate the ability of our framework to generate plausible and practical commercial scenes. © 2023 Owner/Author.

The rapid advances in technologies have brought new interaction paradigms of smart objects (e.g., digital devices) beyond digital device screens. By utilizing spatial properties, configurations, and movements of smart objects, designing spatial interaction, which is one of the emerging interaction paradigms, efficiently promotes engagement with digital content and physical facility. However, as an important phase of design, prototyping such interactions still remains challenging, since there is no ad-hoc approach for this emerging paradigm. Designers usually rely on methods that require fixed hardware setup and advanced coding skills to script and validate early-stage concepts. These requirements restrict the design process to a limited group of users in indoor scenes. To facilitate the prototyping to general usages, we aim to figure out the design difficulties and underlying needs of current design processes for spatially-aware object interactions by empirical studies. Besides, we explore the design space of the spatial interaction for smart objects and discuss the design space in an input-output spatial interaction model. Based on these findings, we present ProObjAR, an all-in-one novel prototyping system with an Augmented Reality Head Mounted Display (AR-HMD). Our system allows designers to easily obtain the spatial data of smart objects being prototyped, specify spatially-aware interactive behaviors from an input-output event triggering workflow, and test the prototyping results in situ. From the user study, we find that ProObjAR simplifies the design procedure and increases design efficiency to a large extent and thus advancing the development of spatially-aware applications in smart ecosystems.

Assembly-based furniture with movable parts enables shape and structure reconfiguration, thus supporting multiple functions. Although a few attempts have been made for facilitating the creation of multi-function objects, designing such a multi-function assembly with the existing solutions often requires high imagination of designers. We develop the Magic Furniture system for users to easily create such designs simply given multiple cross-category objects. Our system automatically leverages the given objects as references to generate a 3D model with movable boards driven by back-and-forth movement mechanisms. By controlling the states of these mechanisms, a designed multi-function furniture object can be reconfigured to approximate the shapes and functions of the given objects. To ensure the designed furniture easy to transform between different functions, we perform an optimization algorithm to choose a proper number of movable boards and determine their shapes and sizes, following a set of design guidelines. We demonstrate the effectiveness of our system through various multi-function furniture designed with different sets of reference inputs and various movement constraints. We also evaluate the design results through several experiments including comparative and user studies. IEEE

Recent works have achieved remarkable performance for action recognition with human skeletal data by utilizing graph convolutional models. Existing models mainly focus on developing graph convolutional operations to encode structural properties of a skeletal graph, whose topology is manually predefined and fixed over all action samples. Some recent works further take sample-dependent relationships among joints into consideration. However, the complex relationships between arbitrary pairwise joints are difficult to learn and the temporal features between frames are not fully exploited by simply using traditional convolutions with small local kernels. In this paper, we propose a motif-based graph convolution method, which makes use of sample-dependent latent relations among non-physically connected joints to impose a high-order locality and assigns different semantic roles to physical neighbors of a joint to encode hierarchical structures. Furthermore, we propose a sparsity-promoting loss function to learn a sparse motif adjacency matrix for latent dependencies in non-physical connections. For extracting effective temporal information, we propose an efficient local temporal block. It adopts partial dense connections to reuse temporal features in local time windows, and enrich a variety of information flow by gradient combination. In addition, we introduce a non-local temporal block to capture global dependencies among frames. Our model can capture local and non-local relationships both spatially and temporally, by integrating the local and non-local temporal blocks into the sparse motif-based graph convolutional networks (SMotif-GCNs). Comprehensive experiments on four large-scale datasets show that our model outperforms the state-of-the-art methods. Our code is publicly available at https://github.com/wenyh1616/SAMotif-GCN.

Maintaining global consistency continues to be critical for online 3D indoor scene reconstruction. However, it is still challenging to generate satisfactory 3D reconstruction in terms of global consistency for previous approaches using purely geometric analysis, even with bundle adjustment or loop closure techniques. In this article, we propose a novel real-time 3D reconstruction approach which effectively integrates both semantic and geometric cues. The key challenge is how to map this indicative information, i.e., semantic priors, into a metric space as measurable information, thus enabling more accurate semantic fusion leveraging both the geometric and semantic cues. To this end, we introduce a semantic space with a continuous metric function measuring the distance between discrete semantic observations. Within the semantic space, we present an accurate frame-to-model semantic tracker for camera pose estimation, and semantic pose graph equipped with semantic links between submaps for globally consistent 3D scene reconstruction. With extensive evaluation on public synthetic and real-world 3D indoor scene RGB-D datasets, we show that our approach outperforms the previous approaches for 3D scene reconstruction both quantitatively and qualitatively, especially in terms of global consistency.

The research topic of sketch-to-portrait generation has witnessed a boost of progress with deep learning techniques. The recently proposed StyleGAN architectures achieve state-of-the-art generation ability but the original StyleGAN is not friendly for sketch-based creation due to its unconditional generation nature. To address this issue, we propose a direct conditioning strategy to better preserve the spatial information under the StyleGAN framework. Specifically, we introduce Spatially Conditioned StyleGAN (SC-StyleGAN for short), which explicitly injects spatial constraints to the original StyleGAN generation process. We explore two input modalities, sketches and semantic maps, which together allow users to express desired generation results more precisely and easily. Based on SC-StyleGAN, we present DrawingInStyles, a novel drawing interface for non-professional users to easily produce high-quality, photo-realistic face images with precise control, either from scratch or editing existing ones. Qualitative and quantitative evaluations show the superior generation ability of our method to existing and alternative solutions. The usability and expressiveness of our system are confirmed by a user study. © 1995-2012 IEEE.

The development of deep generative models has inspired various facial image editing methods, but many of them are difficult to be directly applied to video editing due to various challenges ranging from imposing 3D constraints, preserving identity consistency, ensuring temporal coherence, etc. To address these challenges, we propose a new framework operating on the StyleGAN2 latent space for identity-aware and shape-aware edit propagation on face videos. In order to reduce the difficulties of maintaining the identity, keeping the original 3D motion, and avoiding shape distortions, we disentangle the StyleGAN2 latent vectors of human face video frames to decouple the appearance, shape, expression, and motion from identity. An edit encoding module is used to map a sequence of image frames to continuous latent codes with 3D parametric control and is trained in a self-supervised manner with identity loss and triple shape losses. Our model supports propagation of edits in various forms: I. direct appearance editing on a specific keyframe, II. implicit editing of face shape via a given reference image, and III. existing latent-based semantic edits. Experiments show that our method works well for various forms of videos in the wild and outperforms an animation-based approach and the recent deep generative techniques.

In this work, we present a novel method called WSDesc to learn 3D local descriptors in a weakly supervised manner for robust point cloud registration. Our work builds upon recent 3D CNN-based descriptor extractors, which leverage a voxel-based representation to parameterize local geometry of 3D points. Instead of using a predefined fixed-size local support in voxelization, we propose to learn the optimal support in a data-driven manner. To this end, we design a novel differentiable voxelization layer that can back-propagate the gradient to the support size optimization. To train the extracted descriptors, we propose a novel registration loss based on the deviation from rigidity of 3D transformations, and the loss is weakly supervised by the prior knowledge that the input point clouds have partial overlap, without requiring ground-truth alignment information. Through extensive experiments, we show that our learned descriptors yield superior performance on existing geometric registration benchmarks. © 1995-2012 IEEE.

Vectorizing line drawings is a fundamental component of the workflow in various applications such as graphic design and computer animation. A practical vectorization tool is desired to produce high-quality curves that are faithful to the original inputs and close to the connectivity of human drawings. The existing line vectorization approaches either suffer from low geometry accuracy or incorrect connectivity for noisy inputs or detailed complex drawings. We propose a novel line drawing vectorization framework based on coarse-to-fine curve network optimization. Our technique starts with an initial curve network generated by an existing tracing method. It then performs a global optimization which fits the curve network to image centrelines. Finally, our method performs a finer optimization in local junction regions to achieve better connectivity and curve geometry around junctions. We qualitatively and quantitatively evaluate our system on line drawings with varying image quality and shape complexity, and show that our technique outperforms existing works in terms of curve quality and computational time. © 2023 Eurographics - The European Association for Computer Graphics and John Wiley & Sons Ltd.

We present MobileSky, the first automatic method for real-time high-quality sky replacement for mobile AR applications. The primary challenge of this task is how to extract sky regions in camera feed both quickly and accurately. While the problem of sky replacement is not new, previous methods mainly concern extraction quality rather than efficiency, limiting their application to our task. We aim to provide higher quality, both spatially and temporally consistent sky mask maps for all camera frames in real time. To this end, we develop a novel framework that combines a new deep semantic network called FSNet with novel post-processing refinement steps. By leveraging IMU data, we also propose new sky-aware constraints such as temporal consistency, position consistency, and color consistency to help refine the weakly classified part of the segmentation output. Experiments show that our method achieves an average of around 30 FPS on off-the-shelf smartphones and outperforms the state-of-the-art sky replacement methods in terms of execution speed and quality. In the meantime, our mask maps appear to be visually more stable across frames. Our fast sky replacement method enables several applications, such as AR advertising, art making, generating fantasy celestial objects, visually learning about weather phenomena, and advanced video-based visual effects. To facilitate future research, we also create a new video dataset containing annotated sky regions with IMU data. IEEE

For 3D animators, choreography with artificial intelligence has attracted more attention recently. However, most existing deep learning methods mainly rely on music for dance generation and lack sufficient control over generated dance motions. To address this issue, we introduce the idea of keyframe interpolation for music-driven dance generation and present a novel transition generation technique for choreography. Specifically, this technique synthesizes visually diverse and plausible dance motions by using normalizing flows to learn the probability distribution of dance motions conditioned on a piece of music and a sparse set of key poses. Thus, the generated dance motions respect both the input musical beats and the key poses. To achieve a robust transition of varying lengths between the key poses, we introduce a time embedding at each timestep as an additional condition. Extensive experiments show that our model generates more realistic, diverse, and beat-matching dance motions than the compared state-of-the-art methods, both qualitatively and quantitatively. Our experimental results demonstrate the superiority of the keyframe-based control for improving the diversity of the generated dance motions. IEEE

We propose a novel generative model named as PlanNet for component-based plan synthesis. The proposed model consists of three modules, a wave function collapse algorithm to create large-scale wireframe patterns as the embryonic forms of floor plans, and two deep neural networks to outline the plausible boundary from each squared pattern, and meanwhile estimate the potential semantic labels for the components. In this manner, we use PlanNet to generate a large-scale component-based plan dataset with 10 K examples. Given an input boundary, our method retrieves dataset plan examples with similar configurations to the input, and then transfers the space layout from a user-selected plan example to the input. Benefiting from our interactive workflow, users can recursively subdivide individual components of the plans to enrich the plan contents, thus designing more complex plans for larger scenes. Moreover, our method also adopts a random selection algorithm to make the variations on semantic labels of the plan components, aiming at enriching the 3D scenes that the output plans are suited for. To demonstrate the quality and versatility of our generative model, we conduct intensive experiments, including the analysis of plan examples and their evaluations, plan synthesis with both hard and soft boundary constraints, and 3D scenes designed with the plan subdivision on different scales. We also compare our results with the state-of-the-art floor plan synthesis methods to validate the feasibility and efficacy of the proposed generative model.

Intelligent tools for creating synthetic scenes have been developed significantly in recent years. Existing techniques on interactive scene synthesis only incorporate a single object at every interaction, i.e., crafting a scene through a sequence of single-object insertions with user preferences. These techniques suggest objects by considering existent objects in the scene instead of fully picturing the eventual result, which is inherently problematic since the sets of objects to be inserted are seldom fixed during interactive processes. In this article, we introduce SceneDirector, a novel interactive scene synthesis tool to help users quickly picture various potential synthesis results by simultaneously editing groups of objects. Specifically, groups of objects are rearranged in real-time with respect to a position of an object specified by a mouse cursor or gesture, i.e., a movement of a single object would trigger the rearrangement of the existing object group, the insertions of potentially appropriate objects, and the removal of redundant objects. To achieve this, we first propose an idea of coherent group set which expresses various concepts of layout strategies. Subsequently, we present layout attributes, where users can adjust how objects are arranged by tuning the weights of the attributes. Thus, our method gives users intuitive control of both how to arrange groups of objects and where to place them. Through extensive experiments and two applications, we demonstrate the potentiality of our framework and how it enables concurrently effective and efficient interactions of editing groups of objects.

In this article, we present OrthoAligner, a novel method to predict the visual outcome of orthodontic treatment in a portrait image. Unlike the state-of-the-art method, which relies on a 3D teeth model obtained from dental scanning, our method generates realistic alignment effects in images without requiring additional 3D information as input and thus making our system readily available to average users. The key of our approach is to employ the 3D geometric information encoded in an unsupervised generative model, i.e., StyleGAN in this article. Instead of directly conducting translation in the image space, we embed the teeth region extracted from a given portrait to the latent space of the StyleGAN generator and propose a novel latent editing method to discover a geometrically meaningful editing path that yields the alignment process in the image space. To blend the edited mouth region with the original portrait image, we further introduce a BlendingNet to remove boundary artifacts and correct color inconsistency. We also extend our method to short video clips by propagating the alignment effects across neighboring frames. We evaluate our method in various orthodontic cases, compare it to the state-of-the-art and competitive baselines, and validate the effectiveness of each component. © 1995-2012 IEEE.

Severe haze in Beijing is characterized by the rapid formation of sulfate via the multiphase oxidation of SO2. While many factors, including aerosol oxidants and atmospheric variables, were discovered and investigated, their relative importance remains unclear. Herein, based on the field observation data obtained in Beijing, China, we developed a kinetic model to explore the key factors that determine the multiphase formation of sulfate. Sensitivity tests give the kinetics of each oxidation pathway varying with pH and temperature, based on which the total sulfate formation rate at room temperature (298 K) is calculated to be generally greater than that at standard temperature (273 K), especially during nighttime. Interfacial oxidants are responsible for sulfate formation within a wide pH range, and transition metal ions become more efficient with increased temperature. The multiphase chemistry is additionally affected by aerosol liquid water content (ALWC), particle radius (Rp), and ionic strength (IS). Within the usual aerosol acidity, the kinetic discrepancy induced by different ALWC levels is more significant at the lower temperature, in contrast to the temperature dependence related to Rp, and the effect of IS depends highly on pH. Machine learning reveals the potential importance of temperature, acidity, and Rp. Temperature and acidity are impactful for the formation of both aqueous and interfacial sulfates, whereas Rp only affects the interfacial processes. The discrepancy between nighttime and daytime is considered throughout this study. Overall, this study reveals the key factors for multiphase sulfate formation and is recommended for kinetic evaluation in future laboratory research.

Face reenactment and reconstruction benefit various applications in self-media, VR, etc. Recent face reenactment methods use 2D facial landmarks to implicitly retarget facial expressions and poses from driving videos to source images, while they suffer from pose and expression preservation issues for cross-identity scenarios, i.e., when the source and the driving subjects are different. Current self-supervised face reconstruction methods also demonstrate impressive results. However, these methods do not handle large expressions well, since their training data lacks samples of large expressions, and 2D facial attributes are inaccurate on such samples. To mitigate the above problems, we propose to explore the inner connection between the two tasks, i.e., using face reconstruction to provide sufficient 3D information for reenactment, and synthesizing videos paired with captured face model parameters through face reenactment to enhance the expression module of face reconstruction. In particular, we propose a novel cascade framework named JR2Net for Joint Face Reconstruction and Reenactment, which begins with the training of a coarse reconstruction network, followed by a 3D-aware face reenactment network based on the coarse reconstruction results. In the end, we train an expression tracking network based on our synthesized videos composed by image-face model parameter pairs. Such an expression tracking network can further enhance the coarse face reconstruction. Extensive experiments show that our JR2Net outperforms the state-of-the-art methods on several face reconstruction and reenactment benchmarks. Copyright © 2023, Association for the Advancement of Artificial Intelligence (www.aaai.org).