Automatic furniture layout is long desired for convenient interior design. Leveraging the remarkable visual reasoning capabilities of multimodal large language models (MLLMs), recent methods address layout generation in a static manner, lacking the feedback-driven refinement essential for interactive user engagement. We introduce Chat2Layout, a novel interactive furniture layout generation system that extends the functionality of MLLMs into the realm of interactive layout design. To achieve this, we establish a unified vision-question paradigm for in-context learning, enabling seamless communication with MLLMs to steer their behavior without altering model weights. Within this framework, we present a novel training-free visual prompting mechanism. This involves a visual-text prompting technique that assist MLLMs in reasoning about plausible layout plans, followed by an Offline-to-Online search (O2O-Search) method, which identifies the minimal set of informative references to provide exemplars for visual-text prompting. By employing an agent system with MLLMs as the core controller, we enable bidirectional interaction. The agent not only comprehends the 3D environment and user requirements through linguistic and visual perception but also plans tasks and reasons about actions to generate and arrange furniture within the virtual space. Furthermore, the agent iteratively updates based on visual feedback from execution results. Experimental results demonstrate that our approach facilitates language-interactive generation and arrangement for diverse and complex 3D furniture.

We present GroomLight, a novel method for relightable hair appearance modeling from multi-view images. Existing hair capture methods struggle to balance photorealistic rendering with relighting capabilities. Analytical material models, while physically grounded, often fail to fully capture appearance details. Conversely, neural rendering approaches excel at view synthesis but generalize poorly to novel lighting conditions. GroomLight addresses this challenge by combining the strengths of both paradigms. It employs an extended hair BSDF model to capture primary light transport and a light-aware residual model to reconstruct the remaining details. We further propose a hybrid inverse rendering pipeline to optimize both components, enabling high-fidelity relighting, view synthesis, and material editing. Extensive evaluations on real-world hair data demonstrate state-of-the-art performance of our method. Our project website is at: https://syntec-research.github.io/GroomLight.

Access to high-quality and diverse 3D articulated digital human assets is crucial in various applications, ranging from virtual reality to social platforms. Generative approaches, such as 3D generative adversarial networks (GANs), are rapidly replacing laborious manual content creation tools. Existing 3D GAN frameworks typically rely on explicit representations that are fast to render but offer limited quality, or implicit representations, which offer high capacity but are slow to render, thereby limiting the 3D fidelity in GAN settings. In this work, we introduce layered surface volumes (LSVs) as a new 3D representation for articulatable humans. LSVs represent a human body using multiple textured mesh layers around a conventional mesh template. The explicit mesh layers can be interpreted as a discretized volume with finite thickness located around the template manifold, as such it has the capacity to capture off-template details like hair or accessories, while at the same time it can profit from mesh rasterization for fast rendering. LSVs can be articulated, and they exhibit exceptional efficiency in GAN settings. Trained on unstructured, single-view 2D image datasets, our LSV-GAN generates high-quality and viewconsistent 3D articulated digital humans without the need for view-inconsistent 2D upsampling networks. Project page can be found here.

Despite recent advances in multi-view hair reconstruction, achieving strand-level precision remains a significant challenge due to inherent limitations in existing capture pipelines. We introduce GroomCap , a novel multi-view hair capture method that reconstructs faithful and high-fidelity hair geometry without relying on external data priors. To address the limitations of conventional reconstruction algorithms, we propose a neural implicit representation for hair volume that encodes high-resolution 3D orientation and occupancy from input views. This implicit hair volume is trained with a new volumetric 3D orientation rendering algorithm, coupled with 2D orientation distribution supervision, to effectively prevent the loss of structural information caused by undesired orientation blending. We further propose a Gaussian-based hair optimization strategy to refine the traced hair strands with a novel chained Gaussian representation, utilizing direct photometric supervision from images. Our results demonstrate that GroomCap is able to capture high-quality hair geometries that are not only more precise and detailed than existing methods but also versatile enough for a range of applications.

As a powerful representation of 3D scenes, the neural radiance field (NeRF) enables high-quality novel view synthesis from multi-view images. Stylizing NeRF, however, remains challenging, especially in simulating a text-guided style with both the appearance and the geometry altered simultaneously. In this paper, we present NeRF-Art, a text-guided NeRF stylization approach that manipulates the style of a pre-trained NeRF model with a simple text prompt. Unlike previous approaches that either lack sufficient geometry deformations and texture details or require meshes to guide the stylization, our method can shift a 3D scene to the target style characterized by desired geometry and appearance variations without any mesh guidance. This is achieved by introducing a novel global-local contrastive learning strategy, combined with the directional constraint to simultaneously control both the trajectory and the strength of the target style. Moreover, we adopt a weight regularization method to effectively suppress cloudy artifacts and geometry noises which arise easily when the density field is transformed during geometry stylization. Through extensive experiments on various styles, we demonstrate that our method is effective and robust regarding both single-view stylization quality and cross-view consistency.

Modern 3D-GANs synthesize geometry and texture by training on large-scale datasets with a consistent structure. Training such models on stylized, artistic data, with often unknown, highly variable geometry, and camera information has not yet been shown possible. Can we train a 3D GAN on such artistic data, while maintaining multi-view consistency and texture quality? To this end, we propose an adaptation framework, where the source domain is a pre-trained 3D-GAN, while the target domain is a 2D-GAN trained on artistic datasets. We, then, distill the knowledge from a 2D generator to the source 3D generator. To do that, we first propose an optimization-based method to align the distributions of camera parameters across domains. Second, we propose regularizations necessary to learn high-quality texture, while avoiding degenerate geometric solutions, such as flat shapes. Third, we show a deformation-based technique for modeling exaggerated geometry of artistic domains, enabling-as a byproduct- personalized geometric editing. Finally, we propose a novel inversion method for 3D-GANs linking the latent spaces of the source and the target domains. Our contributions-for the first time-allow for the generation, editing, and animation of personalized artistic 3D avatars on artistic datasets. Project Page: https:/rameenabdal.github.io/3DAvatarGAN

Building animatable and editable models of clothed humans from raw 3D scans and poses is a challenging problem. Existing reposing methods suffer from the limited expressiveness of Linear Blend Skinning (LBS), require costly mesh extraction to generate each new pose, and typically do not preserve surface correspondences across different poses. In this work, we introduce Invertible Neural Skinning (INS) to address these shortcomings. To maintain correspondences, we propose a Pose-conditioned Invertible Network (PIN) architecture, which extends the LBS process by learning additional pose-varying deformations. Next, we combine PIN with a differentiable LBS module to build an expressive and end-to-end Invertible Neural Skinning (INS) pipeline. We demonstrate the strong performance of our method by outperforming the state-of-the-art reposing techniques on clothed humans and preserving surface correspondences, while being an order of magnitude faster. We also perform an ablation study, which shows the usefulness of our pose-conditioning formulation, and our qualitative results display that INS can rectify artefacts introduced by LBS well.

We propose a novel approach for unsupervised 3D animation of non-rigid deformable objects. Our method learns the 3D structure and dynamics of objects solely from single-view RGB videos, and can decompose them into semantically meaningful parts that can be tracked and animated. Using a 3D autodecoder framework, paired with a keypoint estimator via a differentiable PnP algorithm, our model learns the underlying object geometry and parts decomposition in an entirely unsupervised manner. This allows it to perform 3D segmentation, 3D keypoint estimation, novel view synthesis, and animation. We primarily evaluate the framework on two video datasets: VoxCeleb 2562and TEDXPeople 2562. In addition, on the Cats 2562image dataset, we show it even learns compelling 3D geometry from still images. Finally, we show our model can obtain animatable 3D objects from a single or few images11Code and visual results available on our project website: https://snap-research.github.io/unsupervised-volumetric-animation..

Existing 3D-aware image synthesis approaches mainly focus on generating a single canonical object and show limited capacity in composing a complex scene containing a variety of objects. This work presents DisCoScene: a 3D-aware generative model for high-quality and controllable scene synthesis. The key ingredient of our method is a very abstract object-level representation (i.e., 3D bounding boxes without semantic annotation) as the scene layout prior, which is simple to obtain, general to describe various scene contents, and yet informative to disentangle objects and background. Moreover, it serves as an intuitive user control for scene editing. Based on such a prior, the proposed model spatially disentangles the whole scene into object-centric generative radiance fields by learning on only 2D images with the global-local discrimination. Our model obtains the generation fidelity and editing flexibility of individual objects while being able to efficiently compose objects and the background into a complete scene. We demonstrate state-of-the-art performance on many scene datasets, including the challenging Waymo outdoor dataset. Project page can be found here.

Neural implicit fields are powerful for representing 3D scenes and generating high-quality novel views, but it remains challenging to use such implicit representations for creating a 3D human avatar with a specific identity and artistic style that can be easily animated. Our proposed method, AvatarCraft, addresses this challenge by using diffusion models to guide the learning of geometry and texture for a neural avatar based on a single text prompt. We carefully design the optimization framework of neural implicit fields, including a coarse-to-fine multi-bounding box training strategy, shape regularization, and diffusion-based constraints, to produce high-quality geometry and texture. Additionally, we make the human avatar animatable by deforming the neural implicit field with an explicit warping field that maps the target human mesh to a template human mesh, both represented using parametric human models. This simplifies animation and reshaping of the generated avatar by controlling pose and shape parameters. Extensive experiments on various text descriptions show that AvatarCraft is effective and robust in creating human avatars and rendering novel views, poses, and shapes. Our project page is: https://avatar-craft.github.io/.

Toward infinite-scale 3D city synthesis, we propose a novel framework, InfiniCity, which constructs and renders an unconstrainedly large and 3D-grounded environment from random noises. InfiniCity decomposes the seemingly impractical task into three feasible modules, taking advantage of both 2D and 3D data. First, an infinite-pixel image synthesis module generates arbitrary-scale 2D maps from the bird’s-eye view. Next, an octree-based voxel completion module lifts the generated 2D map to 3D octrees. Finally, a voxel-based neural rendering module texturizes the voxels and renders 2D images. InfiniCity can thus synthesize arbitrary-scale and traversable 3D city environments. We quantitatively and qualitatively demonstrate the efficacy of the proposed framework.

Learning the long-range interactions on large-scale mesh-based physical systems with flat Graph Neural Networks (GNNs) and stacking Message Passings (MPs) is challenging due to the scaling complexity w.r.t. the number of nodes and over-smoothing. Therefore, there has been growing interest in the community to introduce multi-scale structures to GNNs for physics simulation. However, current state-of-the-art methods are limited by their reliance on the labor-heavy drawing of coarser meshes or building coarser levels based on spatial proximity, which can introduce wrong edges across geometry boundaries. Inspired by the bipartite graph determination, we propose a novel pooling strategy, bi-stride to tackle the aforementioned limitations. Bi-stride pools nodes on every other frontier of the Breadth-First-Search (BFS), without the need for the manual drawing of coarser meshes and, avoid wrong edges introduced by spatial proximity. Additionally, it enables a reduced number of MP times on each level and the non-parametrized pooling and unpooling by interpolations, similar to convolutional Neural Networks (CNNs), which significantly reduces computational requirements. Experiments show that the proposed framework, BSMS-GNN, significantly outperforms existing methods in terms of both accuracy and computational efficiency in representative physics-based simulation scenarios.

Semantic image synthesis, translating semantic layouts to photo-realistic images, is a one-to-many mapping problem. Though impressive progress has been recently made, diverse semantic synthesis that can efficiently produce semantic-level or even instance-level multimodal results, still remains a challenge. In this article, we propose a novel diverse semantic image synthesis framework from the perspective of semantic class distributions, which naturally supports diverse generation at both semantics and instance level. We achieve this by modeling class-level conditional modulation parameters as continuous probability distributions instead of discrete values, and sampling per-instance modulation parameters through instance-adaptive stochastic sampling that is consistent across the network. Moreover, we propose prior noise remapping, through linear perturbation parameters encoded from paired references, to facilitate supervised training and exemplar-based instance style control at test time. To further extend the user interaction function of the proposed method, we also introduce sketches into the network. In addition, specially designed generator modules, Progressive Growing Module and Multi-Scale Refinement Module, can be used as a general module to improve the performance of complex scene generation. Extensive experiments on multiple datasets show that our method can achieve superior diversity and comparable quality compared to state-of-the-art methods. Codes are available at https://github.com/tzt101/INADE.git.

Despite recent successes in hair acquisition that fits a high-dimensional hair model to a specific input subject, generative hair models, which establish general embedding spaces for encoding, editing, and sampling diverse hairstyles, are way less explored. In this paper, we present GroomGen , the first generative model designed for hair geometry composed of highly-detailed dense strands. Our approach is motivated by two key ideas. First, we construct hair latent spaces covering both individual strands and hairstyles. The latent spaces are compact, expressive, and well-constrained for high-quality and diverse sampling. Second, we adopt a hierarchical hair representation that parameterizes a complete hair model to three levels: single strands, sparse guide hairs, and complete dense hairs. This representation is critical to the compactness of latent spaces, the robustness of training, and the efficiency of inference. Based on this hierarchical latent representation, our proposed pipeline consists of a strand-VAE and a hairstyle-VAE that encode an individual strand and a set of guide hairs to their respective latent spaces, and a hybrid densification step that populates sparse guide hairs to a dense hair model. GroomGen not only enables novel hairstyle sampling and plausible hairstyle interpolation, but also supports interactive editing of complex hairstyles, or can serve as strong data-driven prior for hairstyle reconstruction from images. We demonstrate the superiority of our approach with qualitative examples of diverse sampled hairstyles and quantitative evaluation of generation quality regarding every single component and the entire pipeline.

Exemplar-based portrait stylization is widely attractive and highly desired. Despite recent successes, it remains challenging, especially when considering both texture and geometric styles. In this article, we present the first framework for one-shot 3D portrait style transfer, which can generate 3D face models with both the geometry exaggerated and the texture stylized while preserving the identity from the original content. It requires only one arbitrary style image instead of a large set of training examples for a particular style, provides geometry and texture outputs that are fully parameterized and disentangled, and enables further graphics applications with the 3D representations. The framework consists of two stages. In the first geometric style transfer stage, we use facial landmark translation to capture the coarse geometry style and guide the deformation of the dense 3D face geometry. In the second texture style transfer stage, we focus on performing style transfer on the canonical texture by adopting a differentiable renderer to optimize the texture in a multi-view framework. Experiments show that our method achieves robustly good results on different artistic styles and outperforms existing methods. We also demonstrate the advantages of our method via various 2D and 3D graphics applications.

Face image manipulation via three-dimensional guidance has been widely applied in various interactive scenarios due to its semantically-meaningful understanding and user-friendly controllability. However, existing 3D-morphable-model-based manipulation methods are not directly applicable to out-of-domain faces, such as non-photorealistic paintings, cartoon portraits, or even animals, mainly due to the formidable difficulties in building the model for each specific face domain. To overcome this challenge, we propose, as far as we know, the first method to manipulate faces in arbitrary domains using human 3DMM. This is achieved through two major steps: 1) disentangled mapping from 3DMM parameters to the latent space embedding of a pre-trained StyleGAN2 [1] that guarantees disentangled and precise controls for each semantic attribute; and 2) cross-domain adaptation that bridges domain discrepancies and makes human 3DMM applicable to out-of-domain faces by enforcing a consistent latent space embedding. Experiments and comparisons demonstrate the superiority of our high-quality semantic manipulation method on a variety of face domains with all major 3D facial attributes controllable - pose, expression, shape, albedo, and illumination. Moreover, we develop an intuitive editing interface to support user-friendly control and instant feedback. Our project page is https://cassiepython.github.io/cddfm3d/index.html.

We present CLIP-NeRF, a multi-modal 3D object manipulation method for neural radiance fields (NeRF). By leveraging the joint language-image embedding space of the recent Contrastive Language-Image Pre-Training (CLIP) model, we propose a unified framework that allows manip-ulating NeRF in a user-friendly way, using either a short text prompt or an exemplar image. Specifically, to combine the novel view synthesis capability of NeRF and the controllable manipulation ability of latent representations from generative models, we introduce a disentangled conditional NeRF architecture that allows individual control over both shape and appearance. This is achieved by performing the shape conditioning via applying a learned deformation field to the positional encoding and deferring color conditioning to the volumetric rendering stage. To bridge this disentangled latent representation to the CLIP embedding, we design two code mappers that take a CLIP embedding as input and update the latent codes to reflect the targeted editing. The mappers are trained with a CLIP-based matching loss to ensure the manipulation accuracy. Furthermore, we propose an inverse optimization method that accurately projects an input image to the latent codes for manipulation to enable editing on real images. We evaluate our approach by extensive experiments on a variety of text prompts and exemplar images and also provide an intuitive interface for interactive editing.

Spatially-adaptive normalization (SPADE) is remarkably successful recently in conditional semantic image synthesis in T. Park et al. 2019 which modulates the normalized activation with spatially-varying transformations learned from semantic layouts, to prevent the semantic information from being washed away. Despite its impressive performance, a more thorough understanding of the advantages inside the box is still highly demanded to help reduce the significant computation and parameter overhead introduced by this novel structure. In this paper, from a return-on-investment point of view, we conduct an in-depth analysis of the effectiveness of this spatially-adaptive normalization and observe that its modulation parameters benefit more from semantic-awareness rather than spatial-adaptiveness, especially for high-resolution input masks. Inspired by this observation, we propose class-adaptive normalization (CLADE), a lightweight but equally-effective variant that is only adaptive to semantic class. In order to further improve spatial-adaptiveness, we introduce intra-class positional map encoding calculated from semantic layouts to modulate the normalization parameters of CLADE and propose a truly spatially-adaptive variant of CLADE, namely CLADE-ICPE. Through extensive experiments on multiple challenging datasets, we demonstrate that the proposed CLADE can be generalized to different SPADE-based methods while achieving comparable generation quality compared to SPADE, but it is much more efficient with fewer extra parameters and lower computational cost. The code and pretrained models are available at https://github.com/tzt101/CLADE.git.

We present a novel method to acquire object representations from online image collections, capturing high-quality geometry and material properties of arbitrary objects from photographs with varying cameras, illumination, and backgrounds. This enables various object-centric rendering applications such as novel-view synthesis, relighting, and harmonized background composition from challenging in-the-wild input. Using a multi-stage approach extending neural radiance fields, we first infer the surface geometry and refine the coarsely estimated initial camera parameters, while leveraging coarse foreground object masks to improve the training efficiency and geometry quality. We also introduce a robust normal estimation technique which eliminates the effect of geometric noise while retaining crucial details. Lastly, we extract surface material properties and ambient illumination, represented in spherical harmonics with extensions that handle transient elements, e.g. sharp shadows. The union of these components results in a highly modular and efficient object acquisition framework. Extensive evaluations and comparisons demonstrate the advantages of our approach in capturing high-quality geometry and appearance properties useful for rendering applications.

Traditionally, reduced hair simulation methods are either restricted to heuristic approximations or bound to specific hairstyles. We introduce the first CNN-integrated framework for simulating various hairstyles. The approach produces visually realistic hairs with an interactive speed. To address the technical challenges, our hair simulation pipeline is designed as a two-stage process. First, we present a fully-convolutional neural interpolator as the backbone generator to compute dynamic weights for guide hair interpolation. Then, we adopt a second generator to produce fine-scale displacements to enhance the hair details. We train the neural interpolator with a dedicated loss function and the displacement generator with an adversarial discriminator. Experimental results demonstrate that our method is effective, efficient, and superior to the state-of-the-art on a wide variety of hairstyles. We further propose a performance-driven digital avatar system and an interactive hairstyle editing tool to illustrate the practical applications.

Human motion retargeting aims to transfer the motion of one person in a "driving" video or set of images to another person. Existing efforts leverage a long training video from each target person to train a subject-specific motion transfer model. However, the scalability of such methods is limited, as each model can only generate videos for the given target subject, and such training videos are labor-intensive to acquire and process. Few-shot motion transfer techniques, which only require one or a few images from a target, have recently drawn considerable attention. Methods addressing this task generally use either 2D or explicit 3D representations to transfer motion, and in doing so, sacrifice either accurate geometric modeling or the flexibility of an end-to-end learned representation. Inspired by the Transformable Bottleneck Network, which renders novel views and manipulations of rigid objects, we propose an approach based on an implicit volumetric representation of the image content, which can then be spatially manipulated using volumetric flow fields. We address the challenging question of how to aggregate information across different body poses, learning flow fields that allow for combining content from the appropriate regions of input images of highly non-rigid human subjects performing complex motions into a single implicit volumetric representation. This allows us to learn our 3D representation solely from videos of moving people. Armed with both 3D object understanding and end-to-end learned rendering, this categorically novel representation delivers state-of-the-art image generation quality, as shown by our quantitative and qualitative evaluations.

We propose novel motion representations for animating articulated objects consisting of distinct parts. In a completely unsupervised manner, our method identifies object parts, tracks them in a driving video, and infers their motions by considering their principal axes. In contrast to the previous keypoint-based works, our method extracts meaningful and consistent regions, describing locations, shape, and pose. The regions correspond to semantically relevant and distinct object parts, that are more easily detected in frames of the driving video. To force decoupling of foreground from background, we model non-object related global motion with an additional affine transformation. To facilitate animation and prevent the leakage of the shape of the driving object, we disentangle shape and pose of objects in the region space. Our model1can animate a variety of objects, surpassing previous methods by a large margin on existing benchmarks. We present a challenging new benchmark with high-resolution videos and show that the improvement is particularly pronounced when articulated objects are considered, reaching 96.6% user preference vs. the state of the art.

Semantic image synthesis, translating semantic layouts to photo-realistic images, is a one-to-many mapping problem. Though impressive progress has been recently made, diverse semantic synthesis that can efficiently produce semantic-level multimodal results, still remains a challenge. In this paper, we propose a novel diverse semantic image synthesis framework from the perspective of semantic class distributions, which naturally supports diverse generation at semantic or even instance level. We achieve this by modeling class-level conditional modulation parameters as continuous probability distributions instead of discrete values, and sampling per-instance modulation parameters through instance-adaptive stochastic sampling that is consistent across the network. Moreover, we propose prior noise remapping, through linear perturbation parameters encoded from paired references, to facilitate supervised training and exemplar-based instance style control at test time. Extensive experiments on multiple datasets show that our method can achieve superior diversity and comparable quality compared to state-of-the-art methods. Code will be available at https://github.com/tzt101/INADE.git

Disentangling data into interpretable and independent factors is critical for controllable generation tasks. With the availability of labeled data, supervision can help enforce the separation of specific factors as expected. However, it is often expensive or even impossible to label every single factor to achieve fully-supervised disentanglement. In this paper, we adopt a general setting where all factors that are hard to label or identify are encapsulated as a single unknown factor. Under this setting, we propose a flexible weakly-supervised multi-factor disentanglement framework DisUnknown, which Distills Unknown factors for enabling multi-conditional generation regarding both labeled and unknown factors. Specifically, a two-stage training approach is adopted to first disentangle the unknown factor with an effective and robust training method, and then train the final generator with the proper disentanglement of all labeled factors utilizing the unknown distillation. To demonstrate the generalization capacity and scalability of our method, we evaluate it on multiple benchmark datasets qualitatively and quantitatively and further apply it to various real-world applications on complicated datasets.

Distinguishing aesthetically pleasing food photos from others is an important visual analysis task for social media and ranking systems related to food. Nevertheless, aesthetic assessment of food images remains a challenging and relatively unexplored task, largely due to the lack of related food image datasets and practical knowledge. Thus, we present the Gourmet Photography Dataset (GPD), the first large-scale dataset for aesthetic assessment of food photos. It contains 24,000 images with corresponding binary aesthetic labels, covering a large variety of foods and scenes. We also provide a non-stationary regularization method to combat over-fitting and enhance the ability of tuned models to generalize. Quantitative results from extensive experiments, including a generalization ability test, verify that neural networks trained on the GPD achieve comparable performance to human experts on the task of aesthetic assessment. We reveal several valuable findings to support further research and applications related to visual aesthetic analysis of food images. To encourage further research, we have made the GPD publicly available at https://github.com/Openning07/GPA .

Visual aesthetic assessment has been an active research field for decades. Although latest methods have achieved promising performance on benchmark datasets, they typically rely on a large number of manual annotations including both aesthetic labels and related image attributes. In this paper, we revisit the problem of image aesthetic assessment from the self-supervised feature learning perspective. Our motivation is that a suitable feature representation for image aesthetic assessment should be able to distinguish different expert-designed image manipulations, which have close relationships with negative aesthetic effects. To this end, we design two novel pretext tasks to identify the types and parameters of editing operations applied to synthetic instances. The features from our pretext tasks are then adapted for a one-layer linear classifier to evaluate the performance in terms of binary aesthetic classification. We conduct extensive quantitative experiments on three benchmark datasets and demonstrate that our approach can faithfully extract aesthetics-aware features and outperform alternative pretext schemes. Moreover, we achieve comparable results to state-of-the-art supervised methods that use 10 million labels from ImageNet.

Despite the recent success of face image generation with GANs, conditional hair editing remains challenging due to the under-explored complexity of its geometry and appearance. In this paper, we present MichiGAN (Multi-Input-Conditioned Hair Image GAN), a novel conditional image generation method for interactive portrait hair manipulation. To provide user control over every major hair visual factor, we explicitly disentangle hair into four orthogonal attributes, including shape, structure, appearance, and background. For each of them, we design a corresponding condition module to represent, process, and convert user inputs, and modulate the image generation pipeline in ways that respect the natures of different visual attributes. All these condition modules are integrated with the backbone generator to form the final end-to-end network, which allows fully-conditioned hair generation from multiple user inputs. Upon it, we also build an interactive portrait hair editing system that enables straightforward manipulation of hair by projecting intuitive and high-level user inputs such as painted masks, guiding strokes, or reference photos to well-defined condition representations. Through extensive experiments and evaluations, we demonstrate the superiority of our method regarding both result quality and user controllability.

In this paper, we present a learning-based method to the keyframe-based video stylization that allows an artist to propagate the style from a few selected keyframes to the rest of the sequence. Its key advantage is that the resulting stylization is semantically meaningful, i.e., specific parts of moving objects are stylized according to the artist's intention. In contrast to previous style transfer techniques, our approach does not require any lengthy pre-training process nor a large training dataset. We demonstrate how to train an appearance translation network from scratch using only a few stylized exemplars while implicitly preserving temporal consistency. This leads to a video stylization framework that supports real-time inference, parallel processing, and random access to an arbitrary output frame. It can also merge the content from multiple keyframes without the need to perform an explicit blending operation. We demonstrate its practical utility in various interactive scenarios, where the user paints over a selected keyframe and sees her style transferred to an existing recorded sequence or a live video stream.

Time-lapse videos usually contain visually appealing content but are often difficult and costly to create. In this paper, we present an end-to-end solution to synthesize a time-lapse video from a single outdoor image using deep neural networks. Our key idea is to train a conditional generative adversarial network based on existing datasets of time-lapse videos and image sequences. We propose a multi-frame joint conditional generation framework to effectively learn the correlation between the illumination change of an outdoor scene and the time of the day. We further present a multi-domain training scheme for robust training of our generative models from two datasets with different distributions and missing timestamp labels. Compared to alternative time-lapse video synthesis algorithms, our method uses the timestamp as the control variable and does not require a reference video to guide the synthesis of the final output. We conduct ablation studies to validate our algorithm and compare with state-of-the-art techniques both qualitatively and quantitatively.

Despite the popularity of real-time monocular face tracking systems in many successful applications, one overlooked problem with these systems is rigid instability. It occurs when the input facial motion can be explained by either head pose change or facial expression change, creating ambiguities that often lead to jittery and unstable rigid head poses under large expressions. Existing rigid stabilization methods either employ a heavy anatomically-motivated approach that are unsuitable for real-time applications, or utilize heuristic-based rules that can be problematic under certain expressions. We propose the first rigid stabilization method for real-time monocular face tracking using a dynamic rigidity prior learned from realistic datasets. The prior is defined on a region-based face model and provides dynamic region-based adaptivity for rigid pose optimization during real-time performance. We introduce an effective offline training scheme to learn the dynamic rigidity prior by optimizing the convergence of the rigid pose optimization to the ground-truth poses in the training data. Our real-time face tracking system is an optimization framework that alternates between rigid pose optimization and expression optimization. To ensure tracking accuracy, we combine both robust, drift-free facial landmarks and dense optical flow into the optimization objectives. We evaluate our system extensively against state-of-the-art monocular face tracking systems and achieve significant improvement in tracking accuracy on the high-quality face tracking benchmark. Our system can improve facial-performance-based applications such as facial animation retargeting and virtual face makeup with accurate expression and stable pose. We further validate the dynamic rigidity prior by comparing it against other variants on the tracking accuracy.

We introduce a novel four-view image-based hair modeling method. Given four hair images taken from the front, back, left and right views as input, we first estimate the rough 3D shape of the hair observed in the input using a predefined database of 3D hair models, then synthesize a hair texture on the surface of the shape, from which the hair growing direction information is calculated and used to construct a 3D direction field in the hair volume. Finally, we grow hair strands from the scalp, following the direction field, to produce the 3D hair model, which closely resembles the hair in all input images. Our method does not require that all input images are from the same hair, enabling an effective way to create compelling hair models from images of considerably different hairstyles at different views. We demonstrate the efficacy of our method using a wide range of examples.

Reduced hair models have proven successful for interactively simulating a full head of hair strands, building upon a fundamental assumption that only a small set of guide hairs are needed for explicit simulation, and the rest of the hair move coherently and thus can be interpolated using guide hairs. Unfortunately, hair-solid interactions is a pathological case for traditional reduced hair models, as the motion coherence between hair strands can be arbitrarily broken by interacting with solids. In this paper, we propose an adaptive hair skinning method for interactive hair simulation with hair-solid collisions. We precompute many eligible sets of guide hairs and the corresponding interpolation relationships that are represented using a compact strand-based hair skinning model. At runtime, we simulate only guide hairs; for interpolating every other hair, we adaptively choose its guide hairs, taking into account motion coherence and potential hair-solid collisions. Further, we introduce a two-way collision correction algorithm to allow sparsely sampled guide hairs to resolve collisions with solids that can have small geometric features. Our method enables interactive simulation of more than 150 K hair strands interacting with complex solid objects, using 400 guide hairs. We demonstrate the efficiency and robustness of the method with various hairstyles and user-controlled arbitrary hair-solid interactions.

We introduceAutoHair, the first fully automatic method for 3D hair modeling from a single portrait image, with no user interaction or parameter tuning. Our method efficiently generates complete and high-quality hair geometries, which are comparable to those generated by the state-of-the-art methods, where user interaction is required. The core components of our method are: a novel hierarchical deep neural network for automatic hair segmentation and hair growth direction estimation, trained over an annotated hair image database; and an efficient and automatic data-driven hair matching and modeling algorithm, based on a large set of 3D hair exemplars. We demonstrate the efficacy and robustness of our method on Internet photos, resulting in a database of around 50K 3D hair models and a corresponding hairstyle space that covers a wide variety of real-world hairstyles. We also show novel applications enabled by our method, including 3D hairstyle space navigation and hair-aware image retrieval.

We propose a novel system to reconstruct a high-quality hair depth map from a single portrait photo with minimal user input. We achieve this by combining depth cues such as occlusions, silhouettes, and shading, with a novel 3D helical structural prior for hair reconstruction. We fit a parametric morphable face model to the input photo and construct a base shape in the face, hair and body regions using occlusion and silhouette constraints. We then estimate the normals in the hair region via a Shape-from-Shading-based optimization that uses the lighting inferred from the face model and enforces an adaptive albedo prior that models the typical color and occlusion variations of hair. We introduce a 3D helical hair prior that captures the geometric structure of hair, and show that it can be robustly recovered from the input photo in an automatic manner. Our system combines the base shape, the normals estimated by Shape from Shading, and the 3D helical hair prior to reconstruct high-quality 3D hair models. Our single-image reconstruction closely matches the results of a state-of-the-art multi-view stereo applied on a multi-view stereo dataset. Our technique can reconstruct a wide variety of hairstyles ranging from short to long and from straight to messy, and we demonstrate the use of our 3D hair models for high-quality portrait relighting, novel view synthesis and 3D-printed portrait reliefs.

Realistic hair animation is a crucial component in depicting virtual characters in interactive applications. While much progress has been made in high-quality hair simulation, the overwhelming computation cost hinders similar fidelity in realtime simulations. To bridge this gap, we propose a data-driven solution. Building upon precomputed simulation data, our approach constructs a reduced model to optimally represent hair motion characteristics with a small number of guide hairs and the corresponding interpolation relationships. At runtime, utilizing such a reduced model, we only simulate guide hairs that capture the general hair motion and interpolate all rest strands. We further propose a hair correction method that corrects the resulting hair motion with a position-based model to resolve hair collisions and thus captures motion details. Our hair simulation method enables a simulation of a full head of hairs with over 150K strands in realtime. We demonstrate the efficacy and robustness of our method with various hairstyles and driven motions (e.g., head movement and wind force), and compared against full simulation results that does not appear in the training data.

We present a cone-based ray tracing algorithm for high-quality rendering of furry objects with reflection, refraction and defocus effects. By aggregating many sampling rays in a pixel as a single cone, we significantly reduce the high supersampling rate required by the thin geometry of fur fibers. To reduce the cost of intersecting fur fibers with cones, we construct a bounding volume hierarchy for the fiber geometry to find the fibers potentially intersecting with cones, and use a set of connected ribbons to approximate the projections of these fibers on the image plane. The computational cost of compositing and filtering transparent samples within each cone is effectively reduced by approximating away in-cone variations of shading, opacity and occlusion. The result is a highly efficient ray tracing algorithm for furry objects which is able to render images of quality comparable to those generated by alternative methods, while significantly reducing the rendering time. We demonstrate the rendering quality and performance of our algorithm using several examples and a user study.

Abstract Widely used for morphing between objects with arbitrary topology, distance field interpolation (DFI) handles topological transition naturally without the need for correspondence or remeshing, unlike surface‐based interpolation approaches. However, lack of correspondence in DFI also leads to ineffective control over the morphing process. In particular, unless the user specifies a dense set of landmarks, it is not even possible to measure the distortion of intermediate shapes during interpolation, let alone control it. To remedy such issues, we introduce an approach for establishing correspondence between the interior of two arbitrary objects, formulated as an optimal mass transport problem with a sparse set of landmarks. This correspondence enables us to compute non‐rigid warping functions that better align the source and target objects as well as to incorporate local rigidity constraints to perform as‐rigid‐aspossible DFI. We demonstrate how our approach helps achieve flexible morphing results with a small number of landmarks.

Abstract In this paper we study the problem of hair interpolation: given two 3D hair models, we want to generate a sequence of intermediate hair models that transform from one input to another both smoothly and aesthetically pleasing. We propose an automatic method that efficiently calculates a many‐to‐many strand correspondence between two or more given hair models, taking into account the multi‐scale clustering structure of hair. Experiments demonstrate that hair interpolation can be used for producing more vivid portrait morphing effects and enabling a novel example‐based hair styling methodology, where a user can interactively create new hairstyles by continuously exploring a “style space” spanning multiple input hair models.

This paper presents a single-view hair modeling technique for generating visually and physically plausible 3D hair models with modest user interaction. By solving an unambiguous 3D vector field explicitly from the image and adopting an iterative hair generation algorithm, we can create hair models that not only visually match the original input very well but also possess physical plausibility (e.g., having strand roots fixed on the scalp and preserving the length and continuity of real strands in the image as much as possible). The latter property enables us to manipulate hair in many new ways that were previously very difficult with a single image, such as dynamic simulation or interactive hair shape editing. We further extend the modeling approach to handle simple video input, and generate dynamic 3D hair models. This allows users to manipulate hair in a video or transfer styles from images to videos.

Human hair is known to be very difficult to model or reconstruct. In this paper, we focus on applications related to portrait manipulation and take an application-driven approach to hair modeling. To enable an average user to achieve interesting portrait manipulation results, we develop a single-view hair modeling technique with modest user interaction to meet the unique requirements set by portrait manipulation. Our method relies on heuristics to generate a plausible high-resolution strand-based 3D hair model. This is made possible by an effective high-precision 2D strand tracing algorithm, which explicitly models uncertainty and local layering during tracing. The depth of the traced strands is solved through an optimization, which simultaneously considers depth constraints, layering constraints as well as regularization terms. Our single-view hair modeling enables a number of interesting applications that were previously challenging, including transferring the hairstyle of one subject to another in a potentially different pose, rendering the original portrait in a novel view and image-space hair editing.