Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | https://arxiv.org/abs/2603.06512 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911493483659264 |
|---|---|
| author | Menon, Rohit Mueller-Goldingen, Niklas Pan, Sicong Chenchani, Gokul Krishna Bennewitz, Maren |
| author_facet | Menon, Rohit Mueller-Goldingen, Niklas Pan, Sicong Chenchani, Gokul Krishna Bennewitz, Maren |
| contents | Robotic harvesting in dense crop canopies requires effective interventions that depend not only on geometry, but also on explicit, direction-conditioned relations identifying which organs obstruct a target fruit. We present SG-DOR (Scene Graphs with Direction-Conditioned Occlusion Reasoning), a relational framework that, given instance-segmented organ point clouds, infers a scene graph encoding physical attachments and direction-conditioned occlusion. We introduce an occlusion ranking task for retrieving and ranking candidate leaves for a target fruit and approach direction, and propose a direction-aware graph neural architecture with per-fruit leaf-set attention and union-level aggregation. Experiments on a multi-plant synthetic pepper dataset show improved occlusion prediction (F1=0.73, NDCG@3=0.85) and attachment inference (edge F1=0.83) over strong ablations, yielding a structured relational signal for downstream intervention planning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_06512 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | SG-DOR: Learning Scene Graphs with Direction-Conditioned Occlusion Reasoning for Pepper Plants Menon, Rohit Mueller-Goldingen, Niklas Pan, Sicong Chenchani, Gokul Krishna Bennewitz, Maren Robotics Computer Vision and Pattern Recognition Robotic harvesting in dense crop canopies requires effective interventions that depend not only on geometry, but also on explicit, direction-conditioned relations identifying which organs obstruct a target fruit. We present SG-DOR (Scene Graphs with Direction-Conditioned Occlusion Reasoning), a relational framework that, given instance-segmented organ point clouds, infers a scene graph encoding physical attachments and direction-conditioned occlusion. We introduce an occlusion ranking task for retrieving and ranking candidate leaves for a target fruit and approach direction, and propose a direction-aware graph neural architecture with per-fruit leaf-set attention and union-level aggregation. Experiments on a multi-plant synthetic pepper dataset show improved occlusion prediction (F1=0.73, NDCG@3=0.85) and attachment inference (edge F1=0.83) over strong ablations, yielding a structured relational signal for downstream intervention planning. |
| title | SG-DOR: Learning Scene Graphs with Direction-Conditioned Occlusion Reasoning for Pepper Plants |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.06512 |