Feedback-guided compression improves 3d gaussian splatting storage efficiency
FeCoSplat: Feedback-Guided Compression for Feed-Forward 3D Gaussian Splatting
Computer Vision and Pattern Recognition
Summary
Storing or sending 3D images created by a method called Gaussian Splatting usually takes a lot of space. The researchers found a way to compress these images better by focusing on a middle step in the image process instead of the originals or the final details. They also use feedback from the images they generate to improve compression in two steps. This approach reduces the file size a lot while keeping the final image quality and needing very little computing power to reconstruct the 3D image.
What this means in practice
- •For 3d graphics developers: Compress 3D Gaussian Splatting models more efficiently for faster transmission and storage with minimal reconstruction effort.
- •For augmented reality platform teams: Reduce the data size of 3D views transmitted to devices while keeping good image quality and low decoding complexity.
Authors
Yuxuan Li, Yihang Chen, Yufeng Zhang, Jianfei Cai, Weiyao Lin
Abstract
Feed-forward 3D Gaussian Splatting (3DGS) enables efficient novel-view synthesis from sparse multi-view images, yet its representations remain costly to store and transmit. Existing approaches compress either the input images, incurring heavy receiver-side reconstruction, or the reconstructed Gaussian primitives, which are difficult to compress due to their heterogeneous and irregular attributes. We instead compress compact intermediate features, providing a better balance between compression efficiency and receiver-side complexity. Based on this paradigm, we propose FeCoSplat, a feedback-guided compression framework for feed-forward 3DGS. FeCoSplat first compresses multi-view features to obtain an intermediate 3DGS, whose rendered views are used as feedback to guide a second-stage compression for further refinement. The resulting bitstreams are decoded into a compact implicit state, from which the final Gaussian primitives are reconstructed with a lightweight predictor. Experiments demonstrate that FeCoSplat achieves favorable rate--distortion performance, particularly at low bitrates, while requiring only 3.45M parameters for receiver-side Gaussian reconstruction. Code will be released soon.