FocusGS: Spatial Delta Layers for Local Repair and Deterministic Editing of Trained 3D Gaussian Assets

Researchers introduced FocusGS, a method for local repair and deterministic editing of trained 3D Gaussian Splatting (3DGS) assets, achieving a 7.91 dB PSNR improvement in target regions across 93 evaluation views and a mean edited ROI PSNR of 21.97 dB with a +11.05 dB gain across 83 editing trials. The method, detailed in an arXiv paper (2607.28834v1), uses spatial delta layers and erase-insert factorization to enable precise local maintenance, outperforming text-driven baselines in five public editing cases with 33.17 dB Target-mask PSNR and 0.994 Target-delta Correlation.

arXiv:2607.28834v1 Announce Type: new Abstract: 3D Gaussian Splatting 3DGS is evolving from one-time reconstruction into deliverable, inspectable, and maintainable visual assets. Existing workflows focus on global reconstruction, training-time density control, or open-ended generative editing, leaving trained assets without precise local maintenance. We propose FocusGS, which unifies local repair and deterministic editing as composite spatial deltas. Repair is the purely additive special case: its base-manipulation term is empty, and it adds only local Gaussian bases; deterministic editing uses erase-insert factorization EIF to combine old-carrier erasure with new-content insertion. FocusGS addresses spatial gradient starvation: local repair raises target-region PSNR by 7.91 dB over 93 evaluation views. Across all 83 deterministic editing trials, the target ROI improves, with a trial-averaged mean edited ROI PSNR of 21.97 dB and a mean gain of +11.05 dB; across five public editing cases, FocusGS-EIF reaches 33.17 dB Target-mask PSNR and 0.994 Target-delta Correlation, while both text-driven baselines fail to complete the prescribed updates. FocusGS provides a lightweight, verifiable 3DGS maintenance operator.