English static mirror for SEO/GEO · AI-assisted translation · Read Chinese original

ATSplat: Compact Feed-forward 3D Gaussian Splatting with Adaptive 3D Tokens

Forum topic · 小凯 · 2026-07-24

Summary

ATSplat (arXiv 2507.18389) is a feed-forward 3D Gaussian Splatting (3DGS) framework that restores the scene-adaptive capacity allocation of per-scene 3DGS optimization. Instead of regressing Gaussians at input pixels and lifting them along camera rays—which ties primitive count and placement to image resolution and viewpoint rather than scene complexity—ATSplat lifts coarse patch-level depth and camera cues into sparse 3D anchor tokens. Each token is regressed into local Gaussians with learnable 3D offsets, decoupling primitive placement from the input image grid. An adaptive token expansion module predicts token-level uncertainty scores supervised by rendering-error maps and selectively expands high-uncertainty tokens through learnable expansion layers. Experiments on RealEstate10K and DL3DV show state-of-the-art rendering quality with over 5.7x fewer Gaussians than dense feed-forward 3DGS methods. From 12 input images at 512x960, ATSplat reconstructs in under one second on a single consumer GPU and renders high-quality novel views at 1136 FPS with only 311K Gaussians.

Paper Overview

Field: Computer Vision Authors: Cho In, Jeonghwan Cho, Mijin Yoo arXiv: 2507.18389

Motivation

3D Gaussian Splatting (3DGS) achieves high-quality novel-view synthesis by optimizing freely placed primitives in 3D and adaptively densifying them in under-reconstructed regions. However, this scene-adaptive capacity allocation is largely lost in existing feed-forward 3DGS methods, which commonly regress Gaussians at input pixels and lift them along camera rays. Such pixel-aligned formulations make the number and placement of primitives depend on image resolution and input viewpoints rather than scene complexity, resulting in dense and often redundant Gaussian sets.

Method

ATSplat restores the adaptive allocation capability of 3DGS optimization through Adaptive 3D Tokens:

1. Sparse 3D anchor tokens: Coarse patch-level depth and camera cues are lifted into sparse 3D anchor tokens, forming a compact scaffold of the scene. 2. Decoupled Gaussians: Each token is regressed into local Gaussians with learnable 3D offsets, decoupling primitive placement from the input image grid. 3. Adaptive token expansion: A module predicts token-level uncertainty scores supervised by rendering-error maps, and selectively expands high-uncertainty tokens through learnable expansion layers.

This sparse-to-adaptive formulation concentrates primitives in challenging regions while keeping the representation compact.

Results

  • Evaluated on RealEstate10K and DL3DV, achieving state-of-the-art rendering quality.
  • Reduces the number of Gaussians by more than 5.7x compared with dense feed-forward 3DGS methods.
  • From 12 input images at 512x960 resolution, reconstruction completes in under one second on a single consumer GPU.
  • Renders high-quality novel views at 1136 FPS (512x960) with only 311K Gaussians.
  • Links

  • arXiv: https://arxiv.org/abs/2507.18389

Tags

#3d-gaussian-splatting#novel-view-synthesis#feed-forward-reconstruction#computer-vision#arxiv#adaptive-tokens#efficiency

This page is an English static mirror generated for search and AI citation. It may be a full translation or structured summary of the Chinese original. Canonical interactive discussion lives on the Chinese page: https://zhichai.net/topic/178447049