We present a simple, yet effective diffusion-based method for fine-grained, parametric control over light sources in an image. Existing relighting methods either rely on multiple input views to perform inverse rendering at inference time, or fail to provide explicit control over light changes. Our method fine-tunes a diffusion model on a small set of real raw photograph pairs, supplemented by synthetically rendered images at scale, to elicit its photorealistic prior for the relighting task. We leverage the linearity of light to synthesize image pairs depicting controlled light changes of either a target light source or ambient illumination. Using this data and an appropriate fine-tuning scheme, we train a model for precise illumination changes with explicit control over light intensity and color. Lastly, we show how our method can achieve compelling light editing results, and outperforms existing methods based on user preference.
论文检索
输入标题、作者或关键词，从 306 篇学术成果中精准定位
We present a self-supervised approach to in-the-wild image relighting that enables fully controllable, physically based illumination editing. We achieve this by combining the physical accuracy of traditional rendering with the photorealistic appearance made possible by neural rendering. Our pipeline works by inferring a colored mesh representation of a given scene using monocular estimates of geometry and intrinsic components. This representation allows users to define their desired illumination configuration in 3D. The scene under the new lighting can then be rendered using a path-tracing engine. We send this approximate rendering of the scene through a feed-forward neural renderer to predict the final photorealistic relighting result. We develop a differentiable rendering process to reconstruct in-the-wild scene illumination, enabling self-supervised training of our neural renderer on raw image collections. Our method represents a significant step in bringing the explicit physical control over lights available in typical 3D computer graphics tools, such as Blender, to in-the-wild relighting.
Boundary representation (B-Rep) models serve as the primary representation format in modern CAD systems for describing 3D shapes. While deep learning has achieved success with various geometric representations, B-Reps remain challenging due to their hybrid nature of combining continuous geometry with discrete topological relationships. In this paper, we present Stitch-A-Shape, a B-Rep generation framework that directly models both topology and geometry. This strategy departs from prior work that focuses on either topology or geometry while recovering the other through post-processing. Our method consists of a geometry module that determines the spatial configuration of geometric elements (vertices, curves, and surface control points) and a topology module that establishes connectivity relationships and identifies boundary structures, including outer and inner loops. Our approach leverages a sequential "stitching" representation that mirrors the native data structure and inherent bottom-up organization of B-Rep, assembling geometric entities from vertices through curves to faces. We validate that our framework can handle topological and geometric ambiguities, as well as open surfaces and compound solids. Experiments show that Stitch-A-Shape achieves superior generation quality and computational efficiency compared to existing approaches in unconditional generation tasks, while exhibiting effective capabilities in class-conditional generation and B-Rep autocompletion applications.
The Boundary Representation (B-rep) is a widely used 3D model representation of most consumer products designed with CAD software. However, its highly irregular and sparse set of relationships poses significant challenges for designing a generative model tailored to B-reps. Existing approaches use multi-stage approaches to satisfy the complex constraints sequentially. As a result, the final geometry cannot incorporate user edits due to the non-deterministic dependencies between cascaded stages. In contrast, we propose BrepDiff, a single-stage diffusion model for B-rep generation. We present a masked UV grid representation consisting of structured point samples from faces, serving as input for a diffusion transformer. By introducing an asynchronous and shifted noise schedule, we improve the training signal, enabling the diffusion model to better capture the distribution of UV grids. The explicitness of our masked UV grid representation enables users to intuitively understand and freely design surface geometry without being constrained by topological validity. The interconnectivity can be derived from the face layout, which is later processed into a valid solid volume during post-processing. Our approach achieves performance on par with state-of-the-art cascaded models while offering complex and diverse manipulations of geometry and topology, such as shape completion, merging, and interpolation.
Neural image representations have emerged as a promising approach for encoding and rendering visual data. Combined with learning-based workflows, they demonstrate impressive trade-offs between visual fidelity and memory footprint. Existing methods in this domain, however, often rely on fixed data structures that suboptimally allocate memory or compute-intensive implicit models, hindering their practicality for real-time graphics applications. Inspired by recent advancements in radiance field rendering, we introduce Image-GS, a content-adaptive image representation based on 2D Gaussians. Leveraging a custom differentiable renderer, Image-GS reconstructs images by adaptively allocating and progressively optimizing a group of anisotropic, colored 2D Gaussians. It achieves a favorable balance between visual fidelity and memory efficiency across a variety of stylized images frequently seen in graphics workflows, especially for those showing non-uniformly distributed features and in low-bitrate regimes. Moreover, it supports hardware-friendly rapid random access for real-time usage, requiring only 0.3K MACs to decode a pixel. Through error-guided progressive optimization, Image-GS naturally constructs a smooth level-of-detail hierarchy. We demonstrate its versatility with several applications, including texture compression, semantics-aware compression, and joint image compression and restoration.
3D Gaussian Splatting (3DGS) has advanced radiance field reconstruction by enabling real-time rendering. However, its reliance on Gaussian kernels for geometry and low-order Spherical Harmonics (SH) for color encoding limits its ability to capture complex geometries and diverse colors. We introduce Deformable Beta Splatting (DBS), a deformable and compact approach that enhances both geometry and color representation. DBS replaces Gaussian kernels with deformable Beta Kernels, which offer bounded support and adaptive frequency control to capture fine geometric details with higher fidelity while achieving better memory efficiency. In addition, we extended the Beta Kernel to color encoding, which facilitates improved representation of diffuse and specular components, yielding superior results compared to SH-based methods. Furthermore, Unlike prior densification techniques that depend on Gaussian properties, we mathematically prove that adjusting regularized opacity alone ensures distribution-preserved Markov chain Monte Carlo (MCMC), independent of the splatting kernel type. Experimental results demonstrate that DBS achieves state-of-the-art visual quality while utilizing only 45% of the parameters and rendering 1.5x faster than 3DGS-MCMC, highlighting the superior performance of DBS for real-time radiance field rendering. Interactive demonstrations and source code are available on our project website: https://rongliu-leo.github.io/beta-splatting/.
Virtualized 3D Gaussians: Flexible Cluster-based Level-of-Detail System for Real-Time Rendering of Composed Scenes
PDF ↗3D Gaussian Splatting (3DGS) enables the reconstruction of intricate digital 3D assets from multi-view images by leveraging a set of 3D Gaussian primitives for rendering. Its explicit and discrete representation facilitates the seamless composition of complex digital worlds, offering significant advantages over previous neural implicit methods. However, when applied to large-scale compositions, such as crowd-level scenes, it can encompass numerous 3D Gaussians, posing substantial challenges for real-time rendering. To address this, inspired by Unreal Engine 5’s Nanite system, we propose Virtualized 3D Gaussians (V3DG), a cluster-based LOD solution that constructs hierarchical 3D Gaussian clusters and dynamically selects only the necessary ones to accelerate rendering speed. Our approach consists of two stages: (1) Offline Build, where hierarchical clusters are generated using a local splatting method to minimize visual differences across granularities, and (2) Online Selection, where footprint evaluation determines perceptible clusters for efficient rasterization during rendering. We curate a dataset of synthetic and real-world scenes, including objects, trees, people, and buildings, each requiring 0.1 billion 3D Gaussians to capture fine details. Experiments show that our solution balances rendering efficiency and visual quality across user-defined tolerances, facilitating downstream interactive applications that compose extensive 3DGS assets for consistent rendering performance.
3D Gaussian Splatting (3DGS) has emerged as a mainstream solution for novel view synthesis and 3D reconstruction. By explicitly encoding a 3D scene using a collection of Gaussian kernels, 3DGS achieves high-quality rendering with superior efficiency. As a learning-based approach, 3DGS training has been dealt with the standard stochastic gradient descent (SGD) method, which offers at most linear convergence. Consequently, training often requires tens of minutes, even with GPU acceleration. This paper introduces a (near) second-order convergent training algorithm for 3DGS, leveraging its unique properties. Our approach is inspired by two key observations. First, the attributes of a Gaussian kernel contribute independently to the image-space loss, which endorses isolated and local optimization algorithms. We exploit this by splitting the optimization at the level of individual kernel attributes, analytically constructing small-size Newton systems for each parameter group, and efficiently solving these systems on GPU threads. This achieves Newton-like convergence per training image without relying on the global Hessian. Second, kernels exhibit sparse and structured coupling across input images. This property allows us to effectively utilize spatial information to mitigate overshoot during stochastic training. Our method converges an order faster than standard GPU-based 3DGS training, requiring over 10 × fewer iterations while maintaining or surpassing the quality of the compared with the SGD-based 3DGS reconstructions.
We propose an exact method for embedding a disk-topology triangular mesh onto any convex polygon. The method employs a divide-and-conquer approach, iteratively decomposing the embedding problem into smaller sub-problems that map sub-meshes to convex sub-polygons. The process continues until each triangle in the mesh is naturally embedded into a corresponding 3-sided polygon. The approach is supported by a constructive proof, ensuring its theoretical validity. We translate this proof into a practical algorithm, incorporating various dividing strategies and interpolation weights. Unlike previous methods, our approach preserves the connectivity of the input mesh throughout the embedding process. Extensive experiments demonstrate the efficiency and effectiveness of the proposed method.
We introduce a novel method for directional-field design on meshes, enabling users to specify singularities at any location on a mesh. Our method uses a piecewise power-linear representation for phase and scale, offering precise control over field topology. The resulting fields are smooth and accommodate any singularity index and field symmetry. With this representation, we mitigate the artifacts caused by coarse or uneven meshes. We showcase our approach on meshes with diverse topologies and triangle qualities.
Sketch segmentation involves grouping pixels within a sketch that belong to the same object or instance. It serves as a valuable tool for sketch editing tasks, such as moving, scaling, or removing specific components. While image segmentation models have demonstrated remarkable capabilities in recent years, sketches present unique challenges for these models due to their sparse nature and wide variation in styles. We introduce InkLayer, a method for instance segmentation of raster scene sketches. Our approach adapts state-of-the-art image segmentation and object detection models to the sketch domain by employing class-agnostic fine-tuning and refining segmentation masks using depth cues. Furthermore, our method organizes sketches into sorted layers, where occluded instances are inpainted, enabling advanced sketch editing applications. As existing datasets in this domain lack variation in sketch styles, we construct a synthetic scene sketch segmentation dataset, InkScenes, featuring sketches with diverse brush strokes and varying levels of detail. We use this dataset to demonstrate the robustness of our approach. Code and data for this paper are released at project page: https://inklayer.github.io.
We present the first text-based image editing approach for object parts based on pre-trained diffusion models. Diffusion-based image editing approaches capitalized on the deep understanding of diffusion models of image semantics to perform a variety of edits. However, existing diffusion models lack sufficient understanding of many object parts, hindering fine-grained edits requested by users. To address this, we propose to expand the knowledge of pre-trained diffusion models to allow them to understand various object parts, enabling them to perform fine-grained edits. We achieve this by learning special textual tokens that correspond to different object parts through an efficient token optimization process. These tokens are optimized to produce reliable localization masks at each inference step to localize the editing region. Leveraging these masks, we design feature-blending and adaptive thresholding strategies to execute the edits seamlessly. To evaluate our approach, we establish a benchmark and an evaluation protocol for part editing. Experiments show that our approach outperforms existing editing methods on all metrics and is preferred by users \( 66-90\% \) of the time in conducted user studies.
Despite significant advances in modeling image priors via diffusion models, 3D-aware image editing remains challenging, in part because the object is only specified via a single image. To tackle this challenge, we propose 3D-Fixup, a new framework for editing 2D images guided by learned 3D priors. The framework supports difficult editing situations such as object translation and 3D rotation. To achieve this, we leverage a training-based approach that harnesses the generative power of diffusion models. As video data naturally encodes real-world physical dynamics, we turn to video data for generating training data pairs, i.e., a source and a target frame. Rather than relying solely on a single trained model to infer transformations between source and target frames, we incorporate 3D guidance from an Image-to-3D model, which bridges this challenging task by explicitly projecting 2D information into 3D space. We design a data generation pipeline to ensure high-quality 3D guidance throughout training. Results show that by integrating these 3D priors, 3D-Fixup effectively supports complex, identity coherent 3D-aware edits, achieving high-quality results and advancing the application of diffusion models in realistic image manipulation. The code is provided at https://3dfixup.github.io/.
Image editing is an important task in computer graphics, vision, and VFX, with recent diffusion-based methods achieving fast and high-quality results. However, edits requiring significant structural changes, such as non-rigid deformations, object modifications, or content generation, remain challenging. Existing few step editing approaches produce artifacts such as irrelevant texture or struggle to preserve key attributes of the source image (e.g., pose). We introduce Cora , a novel editing framework that addresses these limitations by introducing correspondence-aware noise correction and interpolated attention maps. Our method aligns textures and structures between the source and target images through semantic correspondence, enabling accurate texture transfer while generating new content when necessary. Cora offers control over the balance between content generation and preservation. Extensive experiments demonstrate that, quantitatively and qualitatively, Cora excels in maintaining structure, textures, and identity across diverse edits, including pose changes, object addition, and texture refinements. User studies confirm that Cora delivers superior results, outperforming alternatives. Project Page & source code: cora-edit.github.io
Generating multiple distinct subjects remains a challenge for existing text-to-image diffusion models. Complex prompts often lead to subject leakage, causing inaccuracies in quantities, attributes, and visual features. Preventing leakage among subjects necessitates knowledge of each subject’s spatial location. Recent methods provide these spatial locations via an external layout control. However, enforcing such a prescribed layout often conflicts with the innate layout dictated by the sampled initial noise, leading to misalignment with the model’s prior. In this work, we introduce a new approach that predicts a spatial layout aligned with the prompt, derived from the initial noise, and refines it throughout the denoising process. By relying on this noise-induced layout, we avoid conflicts with externally imposed layouts and better preserve the model’s prior. Our method employs a small neural network to predict and refine the evolving noise-induced layout at each denoising step, ensuring clear boundaries between subjects while maintaining consistency. Experimental results show that this noise-aligned strategy achieves improved text-image alignment and more stable multi-subject generation compared to existing layout-guided techniques, while preserving the rich diversity of the model’s original distribution.
Gaussian-based human avatars have achieved an unprecedented level of visual fidelity. However, existing approaches based on high-capacity neural networks typically require a desktop GPU to achieve real-time performance for a single avatar, and it remains non-trivial to animate and render such avatars on mobile devices including a standalone VR headset due to substantially limited memory and computational bandwidth. In this paper, we present SqueezeMe, a simple and highly effective framework to convert high-fidelity 3D Gaussian full-body avatars into a lightweight representation that supports both animation and rendering with mobile-grade compute. Our key observation is that the decoding of pose-dependent Gaussian attributes from a neural network creates non-negligible memory and computational overhead. Inspired by blendshapes and linear pose correctives widely used in Computer Graphics, we address this by distilling the pose correctives learned with neural networks into linear layers. Moreover, we further reduce the parameters by sharing the correctives among nearby Gaussians. Combining them with a custom splatting pipeline based on Vulkan, we achieve, for the first time, simultaneous animation and rendering of 3 Gaussian avatars in real-time (72 FPS) on a Meta Quest 3 VR headset.
Generating high-fidelity real-time animated sequences of photorealistic 3D head avatars is important for many graphics applications, including immersive telepresence and movies. This is a challenging problem particularly when rendering digital avatar close-ups for showing character’s facial microfeatures and expressions. To capture the expressive, detailed nature of human heads, including skin furrowing and finer-scale facial movements, we propose to couple locally-defined facial expressions with 3D Gaussian splatting to enable creating ultra-high fidelity, expressive and photorealistic 3D head avatars. In contrast to previous works that operate on a global expression space, we condition our avatar’s dynamics on patch-based local expression features and synthesize 3D Gaussians at a patch level. In particular, we leverage a patch-based geometric 3D face model to extract patch expressions and learn how to translate these into local dynamic skin appearance and motion by coupling the patches with anchor points of Scaffold-GS, a recent hierarchical scene representation. These anchors are then used to synthesize 3D Gaussians on-the-fly, conditioned by patch-expressions and viewing direction. We employ color-based densification and progressive training to obtain high-quality results and faster convergence for high resolution 3K training images. By leveraging patch-level expressions, ScaffoldAvatar consistently achieves state-of-the-art performance with visually natural motion, while encompassing diverse facial expressions and styles in real time.
We propose StreamME, a method focuses on fast 3D avatar reconstruction. The StreamME synchronously records and reconstructs a head avatar from live video streams without any pre-cached data, enabling seamless integration of the reconstructed appearance into downstream applications. This exceptionally fast training strategy, which we refer to as on-the-fly training, is central to our approach. Our method is built upon 3D Gaussian Splatting (3DGS), eliminating the reliance on MLPs in deformable 3DGS and relying solely on geometry, which significantly improves the adaptation speed to facial expression. To further ensure high efficiency in on-the-fly training, we introduced a simplification strategy based on primary points, which distributes the point clouds more sparsely across the facial surface, optimizing points number while maintaining rendering quality. Leveraging the on-the-fly training capabilities, our method protects the facial privacy and reduces communication bandwidth in VR system or online conference. Additionally, it can be directly applied to downstream application such as animation, toonify, and relighting. Please refer to our project page for more details: https://songluchuan.github.io/StreamME/.
Speech-driven 3D facial animation plays a key role in applications such as virtual avatars, gaming, and digital content creation. While existing methods have made significant progress in achieving accurate lip synchronization and generating basic emotional expressions, they often struggle to capture and effectively transfer nuanced performance styles. We propose a novel example-based generation framework that conditions a latent diffusion model on a reference style clip to produce highly expressive and temporally coherent facial animations. To address the challenge of accurately adhering to the style reference, we introduce a novel conditioning mechanism called style basis, which extracts key poses from the reference and additively guides the diffusion generation process to fit the style without compromising lip synchronization quality. This approach enables the model to capture subtle stylistic cues while ensuring that the generated animations align closely with the input speech. Extensive qualitative, quantitative, and perceptual evaluations demonstrate the effectiveness of our method in faithfully reproducing the desired style while achieving superior lip synchronization across various speech scenarios.
We present the first scene-update aerial path planning algorithm specifically designed for detecting and updating change areas in urban environments. While existing methods for large-scale 3D urban scene reconstruction focus on achieving high accuracy and completeness, they are inefficient for scenarios requiring periodic updates, as they often re-explore and reconstruct entire scenes, wasting significant time and resources on unchanged areas. To address this limitation, our method leverages prior reconstructions and change probability statistics to guide UAVs in detecting and focusing on areas likely to have changed. Our approach introduces a novel changeability heuristic to evaluate the likelihood of scene changes, driving the planning of two flight paths: a prior path informed by static priors and a dynamic real-time path that adapts to newly detected changes. Extensive experiments on real-world urban datasets demonstrate that our method significantly reduces flight time and computational overhead while maintaining high-quality updates comparable to full-scene re-exploration and reconstruction. These contributions pave the way for efficient, scalable, and adaptive UAV-based scene updates in complex urban environments.