Squeeze3D: Your 3D Generation Model is Secretly an Extreme Neural Compressor

TL;DR

Squeeze3D introduces a high-ratio 3D data compression framework using pre-trained generative models and implicit priors, enabling efficient encoding and decoding across various 3D formats without requiring real 3D datasets.

Contribution

It presents a novel method that leverages pre-trained 3D models for extremely high compression ratios across multiple 3D data formats, trained solely on synthetic data.

Findings

Achieves up to 2187x compression for textured meshes

Attains 55x compression for point clouds

Maintains visual quality comparable to existing methods

Abstract

We propose Squeeze3D, a novel framework that leverages implicit prior knowledge learnt by existing pre-trained 3D generative models to compress 3D data at extremely high compression ratios. Our approach bridges the latent spaces between a pre-trained encoder and a pre-trained generation model through trainable mapping networks. Any 3D model represented as a mesh, point cloud, or a radiance field is first encoded by the pre-trained encoder and then transformed (i.e. compressed) into a highly compact latent code. This latent code can effectively be used as an extremely compressed representation of the mesh or point cloud. A mapping network transforms the compressed latent code into the latent space of a powerful generative model, which is then conditioned to recreate the original 3D model (i.e. decompression). Squeeze3D is trained entirely on generated synthetic data and does not require any 3D datasets. The Squeeze3D architecture can be flexibly used with existing pre-trained 3D encoders and existing generative models. It can flexibly support different formats, including meshes, point clouds, and radiance fields. Our experiments demonstrate that Squeeze3D achieves compression ratios of up to 2187x for textured meshes, 55x for point clouds, and 619x for radiance fields while maintaining visual quality comparable to many existing methods. Squeeze3D only incurs a small compression and decompression latency since it does not involve training object-specific networks to compress an object.