ZPressor: Bottleneck-Aware Compression for Scalable Feed-Forward 3DGS

TL;DR

ZPressor introduces a scalable, bottleneck-aware compression module for feed-forward 3D Gaussian Splatting models, enabling efficient handling of many input views and improving performance and robustness in novel view synthesis.

Contribution

We propose ZPressor, a lightweight, architecture-agnostic module that compresses multi-view inputs into a compact latent state, significantly enhancing scalability and performance of 3DGS models.

Findings

Enables scaling to over 100 views at 480P resolution on an 80GB GPU.

Improves performance of existing 3DGS models with moderate views.

Enhances robustness under dense view settings on large-scale benchmarks.

Abstract

Feed-forward 3D Gaussian Splatting (3DGS) models have recently emerged as a promising solution for novel view synthesis, enabling one-pass inference without the need for per-scene 3DGS optimization. However, their scalability is fundamentally constrained by the limited capacity of their models, leading to degraded performance or excessive memory consumption as the number of input views increases. In this work, we analyze feed-forward 3DGS frameworks through the lens of the Information Bottleneck principle and introduce ZPressor, a lightweight architecture-agnostic module that enables efficient compression of multi-view inputs into a compact latent state $Z$ that retains essential scene information while discarding redundancy. Concretely, ZPressor enables existing feed-forward 3DGS models to scale to over 100 input views at 480P resolution on an 80GB GPU, by partitioning the views into anchor and support sets and using cross attention to compress the information from the support views into anchor views, forming the compressed latent state $Z$. We show that integrating ZPressor into several state-of-the-art feed-forward 3DGS models consistently improves performance under moderate input views and enhances robustness under dense view settings on two large-scale benchmarks DL3DV-10K and RealEstate10K. The video results, code and trained models are available on our project page: https://lhmd.top/zpressor.