Graphics

Recent advances in 3D Gaussian Splatting (3DGS) present two main directions: feed-forward models offer fast inference in sparse-view settings, while per-scene optimization yields high-quality renderings but is computationally expensive. To combine the benefits of both, we introduce Diff3R, a novel framework that explicitly bridges feed-forward prediction and test-time optimization. By incorporating a differentiable 3DGS optimization layer directly into the training loop, our network learns to predict an optimal initialization for test-time optimization rather than a conventional zero-shot result. To overcome the computational cost of backpropagating through the optimization steps, we propose computing gradients via the Implicit Function Theorem and a scalable, matrix-free PCG solver tailored for 3DGS optimization. Additionally, we incorporate a data-driven uncertainty model into the optimization process by adaptively controlling how much the parameters are allowed to change during optimization. This approach effectively mitigates overfitting in under-constrained regions and increases robustness against input outliers. Since our proposed optimization layer is model-agnostic, we show that it can be seamlessly integrated into existing feed-forward 3DGS architectures for both pose-given and pose-free methods, providing improvements for test-time optimization.

We introduce a new approach to high-fidelity 3D scene reconstruction from multi-view RGB images that tightly couples reconstruction with a strong generative 3D prior. We cast scene reconstruction as conditional 3D generation over a set of spatially-localized, overlapping chunks that together tile the scene, scaling generation to large scene extents. Crucially, we inherit the fidelity and completeness of state-of-the-art generative shape models -- we use Trellis.2 as an example -- which we generalize to the scene level. To this end, we propose a projection-based conditioning mechanism that lifts posed multi-view image features into a coherent 3D representation aligned with the generative model, independent of view ordering and spatially anchored to the scene, yielding high-fidelity, multi-view consistent generated geometry. This enables lifting the strong object-level prior of Trellis.2 to multi-view, scene-scale generation, producing faithful, editable PBR mesh reconstructions of indoor environments. As a result, we obtain high-fidelity results that outperform cutting-edge reconstruction methods by 16%.

We present GuidedSceneGen, a text-to-3D generation framework that produces metrically accurate, globally consistent, and semantically interpretable indoor scenes. Unlike prior text-driven methods that often suffer from geometric drift or scale ambiguity, our approach maintains an absolute world coordinate frame throughout the entire generation process. Starting from a textual scene description, we predict a global 3D layout encoding both semantic and geometric structure, which serves as a guiding proxy for downstream stages. A semantics- and depth-conditioned panoramic diffusion model then synthesizes 360° imagery aligned with the global layout, substantially improving spatial coherence. To explore unobserved regions, we employ a video diffusion model guided by optimized camera trajectories that balances coverage and collision avoidance, achieving up to 10× faster sampling compared to exhaustive path exploration. The generated views are fused using 3D Gaussian Splatting, yielding a consistent and fully navigable 3D scene in absolute scale. GuidedSceneGen enables accurate transfer of object poses and semantic labels from layout to reconstruction, and supports progressive scene expansion without re-alignment. Quantitative results and a user study demonstrate greater 3D consistency and layout plausibility compared to recent panoramic text-to-3D baselines.

CoMe extracts surfaces from Gaussian scene representations using learned confidence values to balance geometric and image-based supervision. Additional regularizers constrain primitive colour and normal variation, while changes to the appearance loss improve consistency. The method focuses on efficient reconstruction in large, unbounded scenes, including cases where view-dependent effects make geometry ambiguous.

The generation of consistent multi-view images is essential for diffusion-based 3D texture synthesis, a key component of modern 3D content pipelines. However, existing methods remain weakly coupled across viewpoints, often exhibiting inconsistencies between views that result in artifacts such as ghosting, misalignment, and texture distortions. We introduce PixelSync, a training-free mechanism that explicitly enforces multi-view consistency during the diffusion process. Our method operates in two stages. First, we compute visibility-aware pixel correspondences between known views in a fast pre-computation step by utilizing geometric information obtained from a given 3D model. Then, during the diffusion process, we enforce consistency through two complementary algorithms: (i) attention synchronization, by modifying the attention scores in the U-Net’s down- and mid-blocks, and (ii) latent synchronization, by progressively aligning features on their corresponding shared pixels. PixelSync can be seamlessly integrated into pre-trained diffusion models without retraining, offering a lightweight and effective solution. We demonstrate the effectiveness of our approach for generative texture synthesis by plugging PixelSync into established multi-view diffusion backbones, showcasing reduced cross-view artifacts and improved global coherence across standard quantitative and qualitative evaluations.

Rendering diffuse global illumination in real‐time is often approximated by pre‐computing and storing irradiance in a 3D grid of probes. As long as most of the scene remains static, probes approximate irradiance for all surfaces immersed in the irradiance volume, including novel dynamic objects. This approach, however, suffers from aliasing artifacts and high memory consumption. We propose Neural Irradiance Volume (NIV), a neural‐based technique that allows accurate real‐time rendering of diffuse global illumination via a compact pre‐computed model, overcoming the limitations of traditional probe‐based methods, such as the expensive memory footprint, aliasing artifacts, and scene‐specific heuristics. The key insight is that neural compression creates an adaptive and amortized representation of irradiance, circumventing the cubic scaling of grid‐based methods. Our superior memory‐scaling improves quality by at least 10x at the same memory budget, and enables a straightforward representation of higher‐dimensional irradiance fields, allowing rendering of time‐varying or dynamic effects without requiring additional computation at runtime. Unlike other neural rendering techniques, our method works within strict realtime constraints, providing fast inference (around 1 ms per frame on consumer GPUs at full HD resolution), reduced memory usage (1–5 MB for medium‐sized scenes), and only requires a G‐buffer as input, without expensive ray tracing or denoising.

Recent advances in text-to-3D scene generation have demonstrated significant potential to transform content creation across multiple industries. Although the research community has made impressive progress in addressing the challenges of this complex task, existing methods often generate environments that are only front-facing, lack visual fidelity, exhibit limited scene understanding, and are typically finetuned for either indoor or outdoor settings. In this work, we address these issues and propose DreamAnywhere, a modular system for the fast generation and prototyping of 3D scenes. Our system synthesizes a 360° panoramic image from text, decomposes it into background and objects, constructs a complete 3D representation through hybrid inpainting, and lifts object masks to detailed 3D objects that are placed in the virtual environment. DreamAnywhere supports immersive navigation and intuitive object-level editing, making it ideal for scene exploration, visual mock-ups, and rapid prototyping, all with minimal manual modeling. These features make our system particularly suitable for low-budget movie production, enabling quick iteration on scene layout and visual tone without the overhead of traditional 3D workflows. Our modular pipeline is highly customizable as it allows components to be replaced independently. Compared to current state-of-the-art text- and image-based 3D scene generation approaches, DreamAnywhere shows significant improvements in coherence in novel view synthesis and achieves competitive image quality, demonstrating its effectiveness across diverse and challenging scenarios. A comprehensive user study demonstrates a clear preference for our method over existing approaches, validating both its technical robustness and practical usefulness.

Sorted Opacity Fields reconstructs meshes from Gaussian scene representations, with particular attention to unbounded environments. Hierarchical sorting and a revised depth formulation make the recovered surface more consistent with the opacity field. Shape regularization and a parallel meshing implementation improve geometric accuracy and reduce extraction time.

This system shares shading work across people viewing the same virtual environment from different perspectives. Surface-based caches allow rendering tasks to be distributed across several GPUs while reusing computed lighting. Instead of sending complete frames between devices, the system transfers selected cache updates, reducing communication as the number of viewers grows.

This method renders objects defined by parametric functions directly on modern GPUs. An adaptive level-of-detail stage subdivides the parameter domain into patches with comparable geometric detail, followed by either hardware tessellation or point-based rasterization. Because the process runs anew for each frame without cached geometry or preprocessing, it also supports animated surfaces while controlling image-space precision.

Although 3D Gaussian Splatting (3DGS) has revolutionized 3D reconstruction, it still faces challenges such as aliasing, projection artifacts, and view inconsistencies, primarily due to the simplification of treating splats as 2D entities. We argue that incorporating full 3D evaluation of Gaussians throughout the 3DGS pipeline can effectively address these issues while preserving rasterization efficiency. Specifically, we introduce an adaptive 3D smoothing filter to mitigate aliasing and present a stable view-space bounding method that eliminates popping artifacts when Gaussians extend beyond the view frustum. Furthermore, we promote tile-based culling to 3D with screen-space planes, accelerating rendering and reducing sorting costs for hierarchical rasterization. Our method achieves state-of-the-art quality on in-distribution evaluation sets and significantly outperforms other approaches for out-of-distribution views. Our qualitative evaluations further demonstrate the effective removal of aliasing, distortions, and popping artifacts, ensuring real-time, artifact-free rendering.

Abstract by the paper’s authors, reproduced under CC BY 4.0.

Rendering fog and other participating media repeatedly for several viewers can duplicate expensive lighting work. This approach caches radiance in a hierarchical, world-space grid of spherical-harmonic probes, allowing both temporal reuse and sharing between viewpoints. The method is designed to improve the cost and quality of multi-view cloud rendering.

3D Gaussian Splatting (3DGS) has rapidly become a leading technique for novel-view synthesis, providing exceptional performance through efficient software-based GPU rasterization. Its versatility enables real-time applications, including on mobile and lower-powered devices. However, 3DGS faces key challenges in virtual reality (VR): (1) temporal artifacts, such as popping during head movements, (2) projection-based distortions that result in disturbing and view-inconsistent floaters, and (3) reduced framerates when rendering large numbers of Gaussians, falling below the critical threshold for VR. Compared to desktop environments, these issues are drastically amplified by large field-of-view, constant head movements, and high resolution of head-mounted displays (HMDs). In this work, we introduce VRSplat: we combine and extend several recent advancements in 3DGS to address challenges of VR holistically. We show how the ideas of Mini-Splatting, StopThePop, and Optimal Projection can complement each other, by modifying the individual techniques and core 3DGS rasterizer. Additionally, we propose an efficient foveated rasterizer that handles focus and peripheral areas in a single GPU launch, avoiding redundant computations and improving GPU utilization. Our method also incorporates a fine-tuning step that optimizes Gaussian parameters based on StopThePop depth evaluations and Optimal Projection. We validate our method through a controlled user study with 25 participants, showing a strong preference for VRSplat over other configurations of Mini-Splatting. VRSplat is the first, systematically evaluated 3DGS approach capable of supporting modern VR applications, achieving 72+ FPS while eliminating popping and stereo-disrupting floaters.

Abstract by the paper’s authors, reproduced under CC BY 4.0.