Papers
arxiv:2606.32036

PointSplat: Compact Gaussian Splatting via Human-Centric Prediction

Published on Jun 30
Authors:
,
,
,
,
,
,
,
,
,

Abstract

PointSplat generates compact 3D human representations by directly predicting Gaussian primitives in 3D space, reducing inter-view redundancy through ray casting and feature fusion.

Producing 3D human representations from input views on the fly is essential for immersive live streaming systems, where representation compactness is as critical as high fidelity given limited computational power and transmission bandwidth. Although recent feed-forward reconstruction methods achieve impressive quality through the view-centric prediction of 3D representations, they repeatedly encode the same subject content across multiple views, leading to significant inter-view redundancy. Our key insight is to perform predictions directly in 3D space, enabling the network to learn and produce a highly compact representation. To this end, we propose PointSplat, a novel human-centric approach that directly infers Gaussian primitives from an input point set. The proposed method first estimates a coarse geometric proxy and performs ray casting to prune redundant points and establish explicit 2D--3D correspondences. Subsequently, it employs a Point-Image Transformer to fuse appearance and geometry features, predicting Gaussian attributes in a single forward pass. This design restricts predictions to foreground regions of interest, substantially reducing the total number of Gaussians while improving novel-view rendering quality. Extensive experiments demonstrate that PointSplat achieves higher efficiency and quality while exhibiting strong robustness to variations in view count and image resolution across multiple datasets.

Community

Sign up or log in to comment

Get this paper in your agent:

hf papers read 2606.32036
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

Cite arxiv.org/abs/2606.32036 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2606.32036 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2606.32036 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.