Blender 5.3 released with Gaussian Splat support
495 points
• 2 days ago
• Article
Link
Blender 5.3 引入了一套重要的全新功能，其中以对 3D Gaussian Splats 的原生支持为核心。用户现在可以从 PLY 、 SPZ 和 glTF 文件导入这些场景，允许将它们作为点云进行操作，通过 Geometry Nodes 进行处理，并在 EEVEE 、 Cycles 和 Workbench 中进行渲染。为促进这一功能，软件包含了用于控制 splat 旋转、缩放和辐射率的新属性，并初步支持使用标准 BSDF 材质对 splats 进行重新打光。
Cycles 的渲染能力得到了扩展，在 Principled BSDF 中加入了光色散，使用户能够在玻璃和宝石中创造逼真的彩虹效果。此外，集成 NVIDIA DLSS 4.5 Ray Reconstruction 提供了强大的新降噪器，在视口中实现接近最终渲染的效果，同时显著减少动画播放过程中的闪烁。 EEVEE 也获得了重大更新，包括新的材质照明节点，赋予艺术家对材质如何响应场景光线的细粒度控制，并支持多种全景相机投影方式，如鱼眼镜头。
工作流程管理方面引入了基于项目的组织方式，允许用户整合相关文件、自定义变量，并为每个项目配套资产库。资产管理进一步改进为能够将资产库直接作为扩展进行共享和安装，可从 Poly Haven 等公共领域来源获取，而无需额外的插件。与此同时，Video Sequencer 也进行了大幅重构，引入了完整的涟漪编辑框架、重写的 blade 工具，并提升了性能，使编辑时保持时间线连续性变得更容易。
用户界面也进行了贴心的优化，例如 Outliner 中支持自定义排序，以及在所有编辑器中搜索侧边栏的能力。界面中集成了 Markdown 支持，尤其体现在全新的 Comment 节点里，帮助更清晰地标注节点树。 Grease Pencil 用户则可以使用新的选择工具、一个强大的 Carver 工具（通过布尔减法在笔触和填充中创建切口），以及同时填充多个区域的能力，从而显著简化动画与绘制流程。
Blender 5.3 introduces a significant suite of new features, headlined by native support for 3D Gaussian Splats. Users can now import these scenes from PLY, SPZ, and glTF files, allowing them to be manipulated as point clouds, processed through Geometry Nodes, and rendered in EEVEE, Cycles, and Workbench. To facilitate this, the software includes new attributes for controlling splat rotation, scale, and radiance, with rudimentary support for relighting splats using standard BSDF materials.
The rendering capabilities in Cycles have been expanded with the addition of light dispersion within the Principled BSDF, enabling users to create realistic rainbow effects in glass and gemstones. Additionally, the integration of NVIDIA DLSS 4.5 Ray Reconstruction provides a powerful new denoiser, offering near-final render quality in the viewport while significantly reducing flickering during animation playback. EEVEE also receives major updates, including new material lighting nodes that grant artists granular control over how materials react to scene lights, alongside support for various panoramic camera projections like fisheye lenses.
Workflow management is addressed with the introduction of project-based organization, allowing users to consolidate related files, custom variables, and per-project asset libraries. Asset management is further improved by the ability to share and install asset libraries directly as extensions, pulling from public domain sources like Poly Haven without needing extra add-ons. Meanwhile, the Video Sequencer has undergone a substantial overhaul, introducing a comprehensive ripple editing framework, a rewritten blade tool, and improved performance, making it much easier to maintain timeline continuity during edits.
The user interface also sees thoughtful refinements, such as custom sorting in the Outliner and the ability to search sidebars across all editors. Markdown support has been integrated into the interface, notably in the new Comment node, which helps in annotating node trees clearly. Grease Pencil users benefit from new selection tools, a powerful Carver tool that uses boolean subtraction to create cuts in strokes and fills, and the ability to fill multiple regions simultaneously, streamlining the animation and drawing process significantly.
75 comments • Comments Link
- Gaussian splatting 仍然是一个备受关注的话题，因为它有望取代或增强传统基于 triangle 的渲染方式，在平滑度与优化方面具备优势。不过，部分技术讨论也指出，诸如 SDFs（Signed Distance Fields，带符号距离场）或"flecks"等替代图元在特定场景中可能更具独特优势。
- 将 Gaussian splatting 集成到 Blender 中，被广泛认为是一个成熟且衔接顺畅的补充，使用户能够在既有生态系统内高效地处理大型数据集。
- 在 splatting 对传统工作流的实用性上，仍存在不少挑战，尤其是对编辑场景中离散对象时较为困难。相比标准的基于 mesh 的建模方式，在重建场景里移动特定的 chair（椅子）仍是一道显著的门槛。
- 3D 内容创作的未来日益与 LLM 驱动的自动化挂钩。在这种趋势下，像 Blender 这样的工具可以通过参数化界面，以及由 AI agents 生成的自定义 geometry nodes 来实现更细致的控制。
- 当前的 AI models 可以很有效地扮演"快速且知识渊博的实习生"，用于场景的 greyboxing 或设置 material nodes，但它们通常缺乏长期、面向任务的学习能力，难以真正替代专业设计师。
- Factory Trained Intelligences（FTIs）依赖静态数据集，这与那些需要把知识固化到特定任务中的系统之间存在差异，而这仍是制约更深入专业应用的重要因素。
- 对于想要进入 Gaussian splatting 领域的人，建议关注专门讲 radiance fields 的 YouTube 频道，以及像 Spirula Studio 这样的专业软件等资源，以便跟上这个快速发展的方向。
- 这个行业显然正朝着一种未来演进：通过 text-to-specification 的工作流自动生成 3D 世界。尽管如此，从入门层面的辅助到专业级的生产之间，仍需要大量人工指导。
这一论述反映出计算机图形学领域正在发生更广泛的转变：传统基于 mesh 的流水线正遭遇来自 Gaussian splatting 等基于点与体素表示方法的挑战。尽管这些新技术在视觉保真度与性能上表现惊人，但在专业生产环境中仍显得力不从心，因为在那样的环境里，模块化编辑和资产控制至关重要。与此同时，将 AI agents 集成到设计软件中，正在把 3D artist 的角色从执行性劳动者转变为参数化系统的导演。即便技术持续进步，也已经出现较为明确的共识：目前的工具更适合作为辅助手段，而非彻底取代人类的专业知识；要实现完全自动化、高清的 3D 生成过渡，在技术可控性与长期模型适配方面仍存在难以消除的障碍。 • Gaussian splatting remains a point of interest for its potential to replace or augment traditional triangle-based rendering, offering smoothness and optimization benefits, though some technical discussions suggest alternative primitives like SDFs (Signed Distance Fields) or "flecks" offer unique advantages for specific applications.
• The integration of Gaussian splatting into Blender is widely viewed as a mature and seamless addition, allowing users to handle large datasets effectively within an established ecosystem.
• Challenges persist regarding the utility of splatting for traditional workflows, specifically the difficulty of editing discrete objects within a scene. Moving a specific chair within a reconstructed scene remains a significant hurdle compared to standard mesh-based modeling.
• The future of 3D content creation is increasingly linked to LLM-driven automation, where tools like Blender can be controlled through parametric interfaces and custom geometry nodes generated by AI agents.
• Current AI models function effectively as fast, knowledgeable interns capable of greyboxing scenes or setting up material nodes, yet they often lack the long-term, on-task learning capacity to fully replace expert designers.
• There is a distinction between Factory Trained Intelligences (FTIs) that rely on static datasets and the need for systems that can crystallize knowledge acquired during specific tasks, which remains a limiting factor for deeper professional application.
• For those looking to enter the Gaussian splatting field, resources such as specific YouTube channels dedicated to radiance fields and specialized software like Spirula Studio are recommended for tracking the rapidly evolving landscape.
• The industry is visibly converging toward a future where 3D worlds are generated via text-to-specification workflows, though the gap between junior-level assistance and professional-grade production still requires significant human guidance.
The discourse reflects a broader shift in computer graphics where traditional mesh-based pipelines are being challenged by point-based and volumetric representations like Gaussian splatting. While these newer technologies offer impressive visual fidelity and performance, they struggle with the requirements of professional production environments where modular editing and asset control are paramount. Concurrently, the integration of AI agents into design software is transforming the role of the 3D artist from a manual laborer to a director of parametric systems. Despite these technical advancements, a clear consensus emerges that current tools are best suited for augmentation rather than total replacement of human expertise, as the transition to fully automated, high-fidelity 3D generation faces lingering hurdles in both technical control and long-term model adaptability.