ConeGS: Error-Guided Densification Using Pixel Cones for Improved Reconstruction with Fewer Primitives

Abstract

3D Gaussian Splatting (3DGS) achieves state-of-the-art image quality and real-time performance in novel view synthesis but often suffers from a suboptimal spatial distribution of primitives. This issue stems from cloning-based densification, which propagates Gaussians along existing geometry, limiting exploration and requiring many primitives to adequately cover the scene.

We present ConeGS, an image-space-informed densification framework that is independent of existing scene geometry state. ConeGS first creates a fast Instant Neural Graphics Primitives (iNGP) reconstruction as a geometric proxy to estimate per-pixel depth. During the subsequent 3DGS optimization, it identifies high-error pixels and inserts new Gaussians along the corresponding viewing cones at the predicted depth values, initializing their size according to the cone diameter. A pre-activation opacity penalty rapidly removes redundant Gaussians, while a primitive budgeting strategy controls the total number of primitives, either by a fixed budget or by adapting to scene complexity, ensuring high reconstruction quality.

Experiments show that ConeGS consistently enhances reconstruction quality and rendering performance across Gaussian budgets, with especially strong gains under tight primitive constraints where efficient placement is crucial.

Overview

Densification using cloning struggles to determine where to add new Gaussians and inherits existing biases, requiring many steps to converge. ConeGS, by contrast, places primitives in areas responsible for high photometric error, guided by a geometric iNGP proxy and sized according to the pixel viewing cone, enabling faster scene integration without reliance on the existing geometry.

Video

Pipeline

Initialization. First, an iNGP reconstruction is obtained to serve as a geometric proxy for object surfaces, guiding the placement of Gaussians both during scene initialization and throughout the 3DGS optimization process.

Optimization. During 3DGS optimization, ConeGS performs error-guided densification by sampling a subset of pixels with high L₁ error. For each sampled pixel, a new Gaussian 𝒢 is created along the pixel's viewing cone at the depth estimated by iNGP and scaled to match the cone's size. New Gaussians are accumulated and, every 100 iterations, inserted into the scene after pruning those with low opacity. Blue arrows indicate gradient updates to Gaussian parameters, and the red arrow marks scene updates.

Results

A more efficient primitive allocation of ConeGS is particularly useful for achieving high reconstruction quality with fewer primitives, enabling real-time, high-quality rendering even with limited resources. In this comparison, the results obtained using our method across various scenes and datasets are superior to those of MCMC (with SfM initialization) and EDGS, which is especially noticeable when the number of primitives is small.

BibTeX


      @inproceedings{baranowski2026conegs, 
        title={ConeGS: Error-Guided Densification Using Pixel Cones for Improved Reconstruction with Fewer Primitives}, 
        author={Bartłomiej Baranowski and Stefano Esposito and Patricia Gschoßmann and Anpei Chen and Andreas Geiger},
        year={2026},
        booktitle = {2026 International Conference on 3D Vision (3DV)}, 
      }