Abstract
LGTM is a feed-forward framework that enables high-fidelity 4K novel view synthesis by predicting compact Gaussian primitives with per-primitive textures, decoupling geometric complexity from rendering resolution.
Existing feed-forward 3D Gaussian Splatting methods predict pixel-aligned primitives, leading to a quadratic growth in primitive count as resolution increases. This fundamentally limits their scalability, making high-resolution synthesis such as 4K intractable. We introduce LGTM (Less Gaussians, Texture More), a feed-forward framework that overcomes this resolution scaling barrier. By predicting compact Gaussian primitives coupled with per-primitive textures, LGTM decouples geometric complexity from rendering resolution. This approach enables high-fidelity 4K novel view synthesis without per-scene optimization, a capability previously out of reach for feed-forward methods, all while using significantly fewer Gaussian primitives. Project page: https://yxlao.github.io/lgtm/
Community
The decoupling of geometric complexity from rendering resolution is a smart approach — similar to what we've seen work in neural radiance fields with feature grids. The per-primitive textures remind me of UV atlas techniques from traditional graphics, but applied to feed-forward networks. One question: does the texture prediction add significant memory overhead during inference? 4K synthesis is impressive, but I'm curious about the VRAM footprint compared to pixel-aligned approaches at the same output resolution.
Get this paper in your agent:
hf papers read 2603.25745 Don't have the latest CLI?
curl -LsSf https://hf.co/cli/install.sh | bash Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper