Variation-aware Flexible 3D Gaussian Editing
This addresses the need for more consistent and flexible 3D editing tools for computer graphics and vision researchers, representing a novel method for a known bottleneck rather than a foundational breakthrough.
The paper tackles the problem of cross-view inconsistencies and limited flexibility in indirect 3D Gaussian editing methods by introducing VF-Editor, which enables native editing of Gaussian primitives through a feedforward variation predictor, achieving improved efficiency and flexibility as validated on public and private datasets.
Indirect editing methods for 3D Gaussian Splatting (3DGS) have recently witnessed significant advancements. These approaches operate by first applying edits in the rendered 2D space and subsequently projecting the modifications back into 3D. However, this paradigm inevitably introduces cross-view inconsistencies and constrains both the flexibility and efficiency of the editing process. To address these challenges, we present VF-Editor, which enables native editing of Gaussian primitives by predicting attribute variations in a feedforward manner. To accurately and efficiently estimate these variations, we design a novel variation predictor distilled from 2D editing knowledge. The predictor encodes the input to generate a variation field and employs two learnable, parallel decoding functions to iteratively infer attribute changes for each 3D Gaussian. Thanks to its unified design, VF-Editor can seamlessly distill editing knowledge from diverse 2D editors and strategies into a single predictor, allowing for flexible and effective knowledge transfer into the 3D domain. Extensive experiments on both public and private datasets reveal the inherent limitations of indirect editing pipelines and validate the effectiveness and flexibility of our approach.