F4Splat: Feed-Forward Predictive Densification for Feed-Forward 3D Gaussian Splatting

2026-03-22

AI Summary

This work addresses the limitations of existing feed-forward 3D Gaussian splatting methods, which employ fixed Gaussian distribution strategies that lead to cross-view redundancy and poor control over the total number of Gaussians. To overcome these issues, we propose a predictive densification strategy that estimates region-level densification scores to capture spatial complexity and multi-view overlap, thereby adaptively allocating Gaussians where needed. Our approach introduces a densification-score-guided mechanism that explicitly controls the total Gaussian count, avoids redundant allocations, and enables flexible trade-offs between reconstruction quality and efficiency without requiring retraining. Experiments demonstrate that our method achieves superior novel view synthesis quality compared to existing uncalibrated feed-forward approaches while significantly reducing the number of Gaussians.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Search and Optimization: Sampling/Simulation-based Search

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Federated recommendation systems and personalization

Abstract

Feed-forward 3D Gaussian Splatting methods enable single-pass reconstruction and real-time rendering. However, they typically adopt rigid pixel-to-Gaussian or voxel-to-Gaussian pipelines that uniformly allocate Gaussians, leading to redundant Gaussians across views. Moreover, they lack an effective mechanism to control the total number of Gaussians while maintaining reconstruction fidelity. To address these limitations, we present F4Splat, which performs Feed-Forward predictive densification for Feed-Forward 3D Gaussian Splatting, introducing a densification-score-guided allocation strategy that adaptively distributes Gaussians according to spatial complexity and multi-view overlap. Our model predicts per-region densification scores to estimate the required Gaussian density and allows explicit control over the final Gaussian budget without retraining. This spatially adaptive allocation reduces redundancy in simple regions and minimizes duplicate Gaussians across overlapping views, producing compact yet high-quality 3D representations. Extensive experiments demonstrate that our model achieves superior novel-view synthesis performance compared to prior uncalibrated feed-forward methods, while using significantly fewer Gaussians.

Problem

• 3D Gaussian Splatting
• feed-forward reconstruction
• Gaussian redundancy
• densification control
• novel-view synthesis

Innovation

• predictive densification
• adaptive Gaussian allocation
• feed-forward 3D Gaussian Splatting
• densification score
• compact 3D representation