# gaussian splatting

Published articles for gaussian splatting.

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## Опенсорсим YaGS Plugin: открытый плагин для работы с 3DGS внутри Unreal Engine

DevFeed: [Опенсорсим YaGS Plugin: открытый плагин для работы с 3DGS внутри Unreal Engine](<https://devfeed.tech/articles/yags-plugin-3dgs-unreal-engine-24862.md>)

Original publisher: [Read original article](<https://habr.com/ru/companies/yandex/articles/1052328/>)

Author: s\_eliseev (Яндекс)

Published: 2026-07-02T07:02:10Z

Content type: article

Language: ru

Sources: [Яндекс - Как мы делаем Яндекс / Статьи](<https://devfeed.tech/sources/source.md>)

Topics: [3D](<https://devfeed.tech/topics/3d.md>), [Unreal Engine](<https://devfeed.tech/topics/unreal-engine.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [3dgs](<https://devfeed.tech/tags/3dgs.md>), [gaussian](<https://devfeed.tech/tags/gaussian.md>), [gaussian-splatting](<https://devfeed.tech/tags/gaussian-splatting.md>), [plugin](<https://devfeed.tech/tags/plugin.md>), [tag-831b63de9433](<https://devfeed.tech/tags/tag-831b63de9433.md>), [tag-9e9e5e80d07a](<https://devfeed.tech/tags/tag-9e9e5e80d07a.md>), [tag-e6d9cc1f0757](<https://devfeed.tech/tags/tag-e6d9cc1f0757.md>), [unreal-engine](<https://devfeed.tech/tags/unreal-engine.md>)

### AI overview

The article explains how Gaussian Splatting can create photorealistic 3D representations from photographs for film production, city digital twins, and autonomous-transport simulation. It introduces YaGS Plugin, an open-source plugin that integrates Gaussian Splatting scenes into Unreal Engine, and describes the underlying 3DGS representation and reconstruction process.

### Source excerpt

Представьте, что вам нужна фотореалистичная 3D-копия реального объекта или сцены: например, лица актёра для крупного плана или целой городской улицы. Сегодня для этого есть два инструмента: классическое 3D-моделирование и фотограмметрия. Однако они съедают недели работы и ломаются там, где очень важны детали, -- волосы, хром, стекло, отражения превращаются в артефакты. А на то, чтобы снять реалистичную сцену или объект так, чтобы это масштабировалось на тысячи единиц, обычно уходит слишком много времени и денег. Выход из этого тупика для целого ряда задач мы нашли в Gaussian Splatting и прогнали технологию через несколько очень разных проектов: сканирование людей и локаций для кино (например, для "Кибердеревни"), создание цифровых двойников городских кварталов и симуляцию для автономного транспорта. Инструмент один -- задачи разные. И именно на этом контрасте и родилось самое интересное. Сегодня я расскажу, как риг из 15 камер заменяет установку за сотни тысяч долларов и зачем мы используем движущегося человека как калибровочную доску. Объясню, почему для беспилотного автомобиля фотореализм обязан рождаться из данных, а не из 3D-редактора. Ну и, конечно же, расскажу про YaGS Plugin -- наш опенсорс-плагин, который делает Gaussian Splatting полноценной частью сцены в Unreal Engine. Читать далее

## MELON: Reconstructing 3D objects from images with unknown poses

DevFeed: [MELON: Reconstructing 3D objects from images with unknown poses](<https://devfeed.tech/articles/melon-reconstructing-3d-objects-from-images-with-unknown-poses-28561.md>)

Original publisher: [Read original article](<http://blog.research.google/2024/03/melon-reconstructing-3d-objects-from.html>)

Author: Google AI (noreply@blogger.com)

Published: 2024-03-18T18:41:00Z

Content type: article

Language: en

Sources: [Google Research](<https://devfeed.tech/sources/google-research.md>)

Topics: [3D](<https://devfeed.tech/topics/3d.md>), [Computer vision](<https://devfeed.tech/topics/computer-vision.md>), [Inference](<https://devfeed.tech/topics/inference.md>), [Google](<https://devfeed.tech/topics/google.md>)

Tags: [3d](<https://devfeed.tech/tags/3d.md>), [computer-vision](<https://devfeed.tech/tags/computer-vision.md>), [gans](<https://devfeed.tech/tags/gans.md>), [gaussian-splatting](<https://devfeed.tech/tags/gaussian-splatting.md>), [generative](<https://devfeed.tech/tags/generative.md>), [google](<https://devfeed.tech/tags/google.md>), [images](<https://devfeed.tech/tags/images.md>), [inference](<https://devfeed.tech/tags/inference.md>), [machine-learning](<https://devfeed.tech/tags/machine-learning.md>), [models](<https://devfeed.tech/tags/models.md>), [neural](<https://devfeed.tech/tags/neural.md>), [research](<https://devfeed.tech/tags/research.md>), [rgb](<https://devfeed.tech/tags/rgb.md>), [rotation](<https://devfeed.tech/tags/rotation.md>), [techniques](<https://devfeed.tech/tags/techniques.md>), [vision](<https://devfeed.tech/tags/vision.md>)

### AI overview

This Google Research article explains the challenge of reconstructing 3D objects from a small number of images when the camera poses are unknown. It covers pose inference, pseudo-symmetries, local-minimum failures, and prior approaches including NeRF, 3D Gaussian Splatting, GAN-based methods, BARF, SAMURAI, GNeRF, VMRF, SparsePose, and RUST.

### Source excerpt

Posted by Mark Matthews, Senior Software Engineer, and Dmitry Lagun, Research Scientist, Google Research A person's prior experience and understanding of the world generally enables them to easily infer what an object looks like in whole, even if only looking at a few 2D pictures of it. Yet the capacity for a computer to reconstruct the shape of an object in 3D given only a few images has remained a difficult algorithmic problem for years. This fundamental computer vision task has applications ranging from the creation of e-commerce 3D models to autonomous vehicle navigation. A key part of the problem is how to determine the exact positions from which images were taken, known as pose inference. If camera poses are known, a range of successful techniques -- such as neural radiance fields (NeRF) or 3D Gaussian Splatting -- can reconstruct an object in 3D. But if these poses are not available, then we face a difficult "chicken and egg" problem where we could determine the poses if we knew the 3D object, but we can't reconstruct the 3D object until we know the camera poses. The problem is made harder by pseudo-symmetries -- i.e., many objects look similar when viewed from different angles. For example, square objects like a chair tend to look similar every 90° rotation. Pseudo-symmetries of an object can be revealed by rendering it on a turntable from various angles and plotting its photometric self-similarity map. Self-Similarity map of a toy truck model. Left: The model is rendered on a turntable from various azimuthal angles, θ. Right: The average L2 RGB similarity of a rendering from θ with that of θ*. The pseudo-similarities are indicated by the dashed red lines. The diagram above only visualizes one dimension of rotation. It becomes even more complex (and difficult to visualize) when introducing more degrees of freedom. Pseudo-symmetries make the problem ill-posed, with naïve approaches often converging to local minima. In practice, such an approach might mistake the