FurnSet: Exploiting Repeats for 3D Scene Reconstruction
This addresses 3D scene reconstruction from single images, an incremental improvement for computer vision applications by explicitly using repeated objects.
The paper tackles single-view 3D scene reconstruction by exploiting repeated object instances, proposing FurnSet with per-object tokens and set-aware attention to group identical instances for joint reconstruction. Experiments on 3D-Future and 3D-Front datasets show improved reconstruction quality, demonstrating the effectiveness of leveraging repetition.
Single-view 3D scene reconstruction involves inferring both object geometry and spatial layout. Existing methods typically reconstruct objects independently or rely on implicit scene context, failing to exploit the repeated instances commonly present in realworld scenes. We propose FurnSet, a framework that explicitly identifies and leverages repeated object instances to improve reconstruction. Our method introduces per-object CLS tokens and a set-aware self-attention mechanism that groups identical instances and aggregates complementary observations across them, enabling joint reconstruction. We further combine scene-level and object-level conditioning to guide object reconstruction, followed by layout optimization using object point clouds with 3D and 2D projection losses for scene alignment. Experiments on 3D-Future and 3D-Front demonstrate improved scene reconstruction quality, highlighting the effectiveness of exploiting repetition for robust 3D scene reconstruction.