3D Gaussian Splatting for Real-Time Radiance Field Rendering

Published on Aug 8
· Featured in Daily Papers on Aug 9


Radiance Field methods have recently revolutionized novel-view synthesis of scenes captured with multiple photos or videos. However, achieving high visual quality still requires neural networks that are costly to train and render, while recent faster methods inevitably trade off speed for quality. For unbounded and complete scenes (rather than isolated objects) and 1080p resolution rendering, no current method can achieve real-time display rates. We introduce three key elements that allow us to achieve state-of-the-art visual quality while maintaining competitive training times and importantly allow high-quality real-time (>= 30 fps) novel-view synthesis at 1080p resolution. First, starting from sparse points produced during camera calibration, we represent the scene with 3D Gaussians that preserve desirable properties of continuous volumetric radiance fields for scene optimization while avoiding unnecessary computation in empty space; Second, we perform interleaved optimization/density control of the 3D Gaussians, notably optimizing anisotropic covariance to achieve an accurate representation of the scene; Third, we develop a fast visibility-aware rendering algorithm that supports anisotropic splatting and both accelerates training and allows realtime rendering. We demonstrate state-of-the-art visual quality and real-time rendering on several established datasets.

View arXiv page View PDF


This comment has been hidden

where and how can we access the Model?

Thats great!

Il s'agit d'un véhicule fait avec MidJourney, voilà le prompt que j'ai utilisé :

Lots of people in line wander around behind huge very rusty garbage truck with bull bars, big chain in the front, spikes on the bull bars, dirty windows, the truck spits out huge black smoke, Madmax style, giger style. The truck is in the desert and is moving on a muddy road, realistic

Je l'ai mis ici pour en faire une scène 3D mais je ne sais pas comment faire.

@chabgyver C'est seulement un "paper" pour l'instant, il n'y a pas encore un "space" disponible pour tester. Parfois, tu peux trouver un peu avant HuggingFace le github et tester le code avec google colab par exemple sous la forme d'un notebook/jupyter. Si ce que je t'écris ressemble à du chinois, colle ma prose dans ChatGPT et tu vas piger en 3 sec c'est très simple...

J'ai bien compris ton message, pas grave, j'attendrai que cette technologie soit plus aboutie mais j'ai hâte de tester sur mes propres tableaux

This looks incredible.

Sign up or log in to comment

Models citing this paper 0

No model linking this paper

Cite in a model to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite in a dataset to link it from this page.

Spaces citing this paper 1