AI Paper Atlas

Research map / 3D vision / Gaussian splatting

Beyond Heuristics: Learnable Density Control for 3D Gaussian Splatting

Zhenhua Ning, Xin Li, Jun Yu, Guangming Lu, Yaowei Wang, Wenjie Pei

ICML 2026 · 3D vision › Gaussian splatting · approach: Compact and efficient Gaussian splatting · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add learnable densification (LD) policy to FastGS* (ablation)FastGS* · Mip-NeRF 360PSNR: 27.73 -> 28.14Improved
add learnable pruning (LP) on top of learnable densification (ablation)FastGS* + LD · Mip-NeRF 360PSNR / #Gaussians: 28.14 -> 28.30 PSNR; 2.48 -> 2.17 M GaussiansImproved
add sensitivity-based reward instead of rendering-weight reward (ablation)LeGS · Mip-NeRF 360PSNR: 28.01 -> 28.30Improved
use maintain-action value instead of PPO critic network (ablation)LeGS · Mip-NeRF 360number of Gaussians (M): 2.63 -> 2.17Improved
use maintain-action value instead of PPO critic network (ablation)LeGS · Mip-NeRF 360PSNR: 28.17 -> 28.30No clear effect
closed-form sensitivity score instead of naive re-rendering (ablation)LeGS reward computation · Mip-NeRF 360 (Bicycle)extra computing time (ms): 30.5 -> 0.9Improved
learned RL policy instead of FastGS heuristic density controlFastGS backend · Mip-NeRF 360PSNR: 27.87 -> 28.30Improved
learned RL policy instead of FastGS heuristic density controlFastGS backend · Deep BlendingPSNR: 30.17 -> 30.35Improved
increase FastGS Gaussian count via lower gradient threshold (FastGS*)FastGS · Mip-NeRF 360PSNR: 27.87 -> 27.73Hurt
learned policy vs heuristic: proportion of locally optimal actionsLeGS vs FastGS · Mip-NeRF 360, iteration 15000locally optimal action rate (%): 35.60% -> 64.85%Improved
add RL policy training and sensitivity reward to FastGS*FastGS* · Mip-NeRF 360total training time (s): 599 -> 745Hurt
LeGS vs FastGS* peak GPU memoryLeGS · Mip-NeRF 360peak GPU memory (GB): 8.99 -> 12.57Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy