Research map / Point clouds / Segmentation and detection
Three-Dimensional Object Detection and Layout Prediction Using Clouds of Oriented Gradients
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add COG appearance features to geometric features (ablation) | 3D cuboid detector · SUN RGB-D object categories | average precision: 8.29 -> 52.98 | Helped |
| Use pose-aligned COG bins instead of fixed 2D HOG bins (ablation) | Geom+appearance detector · SUN RGB-D object detection | detection performance | Helped |
| Add cascaded contextual classification (ablation) | Geom+COG detector · SUN RGB-D object categories | average precision: 52.98 -> 58.72 | Helped |
| Use context from 10 categories rather than 5 (ablation) | Cascaded object detector · SUN RGB-D bed | average precision: 58.72 -> 61.29 | Helped |
| Add layout context to the 10-category cascade (ablation) | Cascaded object detector · SUN RGB-D bed | average precision: 61.29 -> 63.67 | Helped |
| Replace standard cubical voxels with Manhattan voxels (ablation) | 3D room layout predictor · SUN RGB-D room layout | free-space IoU: 72.33 -> 78.96 | Helped |
| Add contextual cues from detected objects (ablation) | Manhattan voxel layout predictor · SUN RGB-D room layout | free-space IoU: 78.96 -> 80.23 | Helped |
| Use the proposed COG, Manhattan voxel, and context system instead of Sliding-Shape+Plane-Fitting | Total scene understanding · SUN RGB-D | IoU: 66.0 -> 72.0 | Helped |
| Use COG+context detector instead of Sliding-Shape | SUN RGB-D object detector · SUN RGB-D bed | average precision: 42.95 -> 63.67 | Helped |
Closest papers by meaning
- Cubify Anything: Scaling Indoor 3D Object Detection (CVPR 2025 · similarity 0.89)
- Amodal Detection of 3D Objects: Inferring 3D Bounding Boxes From 2D Ones in RGB-Depth Images (CVPR 2017 · similarity 0.88)
- Uncertainty Meets Diversity: A Comprehensive Active Learning Framework for Indoor 3D Object Detection (CVPR 2025 · similarity 0.87)
- Rooms from Motion: Un-posed Indoor 3D Object Detection as Localization and Mapping (NeurIPS 2025 · similarity 0.87)
- UniDet3D: Multi-dataset Indoor 3D Object Detection (AAAI 2025 · similarity 0.86)
- SPGroup3D: Superpoint Grouping Network for Indoor 3D Object Detection (AAAI 2024 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.