Research map / Detection and segmentation / 3D detection and driving
Augmented Autoencoders: Implicit 3D Orientation Learning for 6D Object Detection
Martin Sundermeyer, Zoltán-Csaba Márton, Maximilian Durner, Rudolph Triebel
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 perspective correction of rotation estimate (ablation) | AAE RetinaNet RGB · T-LESS recall | object recall: 18.35 -> 19.26 (+0.91) | Improved |
| add random add-brightness augmentation (ablation) | AAE · T-LESS obj 5 Primesense | AUC_vsd: 0.472 -> 0.611 | Improved |
| add contrast augmentation (ablation) | AAE · T-LESS obj 5 Primesense | AUC_vsd: 0.611 -> 0.825 | Improved |
| add multiply augmentation (ablation) | AAE · T-LESS obj 5 Primesense | AUC_vsd: 0.825 -> 0.876 | Improved |
| add colour channel inversion (ablation) | AAE · T-LESS obj 5 Primesense | AUC_vsd: 0.876 -> 0.877 | No clear effect |
| add contrast augmentation (ablation) | AAE · T-LESS obj 5 Kinect | AUC_vsd: 0.580 -> 0.701 | Improved |
| add multiply augmentation (ablation) | AAE · T-LESS obj 5 Kinect | AUC_vsd: 0.701 -> 0.855 | Improved |
| add colour channel inversion (ablation) | AAE · T-LESS obj 5 Kinect | AUC_vsd: 0.855 -> 0.897 | Improved |
| add random-position dynamic light (ablation) | AAE · T-LESS obj 5 Primesense | AUC_vsd: 0.861 -> 0.877 | Improved |
| add random-position dynamic light (ablation) | AAE · T-LESS obj 5 Kinect | AUC_vsd: 0.903 -> 0.897 | Hurt |
| train on real Primesense recordings instead of synthetic (ablation) | AAE · T-LESS obj 5 Primesense | AUC_vsd: 0.877 -> 0.890 | No clear effect |
| train on CAD models instead of 3D reconstructions (ablation) | AAE RetinaNet · T-LESS recall | object recall: 19.26 -> 26.79 | Improved |
Closest papers by meaning
- Implicit 3D Orientation Learning for 6D Object Detection from RGB Images (ECCV 2018 · similarity 0.99)
- FS-Net: Fast Shape-Based Network for Category-Level 6D Object Pose Estimation With Decoupled Rotation Mechanism (CVPR 2021 · similarity 0.88)
- Deep Learning of Local RGB-D Patches for 3D Object Detection and 6D Pose Estimation (ECCV 2016 · similarity 0.88)
- Omni6D: Large-Vocabulary 3D Object Dataset for Category-Level 6D Object Pose Estimation (ECCV 2024 · similarity 0.88)
- Pose without Guesses: Generalizable Object Pose Estimation From a Single Reference (IJCV 2026 · similarity 0.88)
- Pos3R: 6D Pose Estimation for Unseen Objects Made Easy (CVPR 2025 · similarity 0.87)
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.
