Search papers, labs, and topics across Lattice.
Seed2GS introduces a novel approach for extracting target objects from pre-built 3D Gaussian Splatting scenes without the need for original reconstruction cameras or extensive training. By decoupling target identity from 3D coverage, the method utilizes QD-SAM3 to select a reliable reference mask and employs visibility-adaptive techniques to reveal objects from various viewpoints. Achieving a mean intersection over union (mIoU) of 92.1% on LERF-MASK, Seed2GS outperforms existing scene-trained and camera-free baselines, demonstrating its efficiency and accuracy in interactive 3D editing.
Achieving 92.1% mIoU without training or original cameras, Seed2GS revolutionizes object extraction from 3D scenes.
Extracting a target object from a pre-built 3D Gaussian Splatting (3DGS) scene enables interactive 3D editing. Existing methods either train for tens of minutes per scene, sacrifice accuracy, or require original reconstruction cameras that pre-built assets may not include. We present Seed2GS, which achieves the highest reported LERF-MASK accuracy without original reconstruction cameras or scene-specific representation training. Its key insight is to separate target identity from 3D coverage. QD-SAM3 selects one reliable reference mask from several open-vocabulary candidates, fixing identity once. Seed lift and visibility-adaptive virtual orbits then expose the object from new viewpoints, while tracking propagates the seed without repeated detection. Because the scene remains frozen, these masks supervise only one temporary foreground logit per Gaussian. On LERF-MASK, Seed2GS reaches 92.1% mean intersection over union (mIoU) with a measured compute-only latency of 9.3 seconds, 3.7 points above the strongest scene-trained baseline and 7.6 points above the closest camera-free baseline. With one fixed test reference per scene, the complete pipeline retains 91.1% mIoU; replacing its predicted seed with a ground-truth mask improves mIoU by only 0.72 points. On 3D-OVS, Seed2GS reaches 95.7% mIoU.
