Research map / Detection and segmentation / CNN features
RoomNet: End-To-End Room Layout Estimation
Chen-Yu Lee, Vijay Badrinarayanan, Tomasz Malisiewicz, Andrew Rabinovich
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 recurrent keypoint refinement (3 iterations) (ablation) | RoomNet basic · LSUN | keypoint error (%): 6.95 -> 6.30 | Improved |
| add recurrent keypoint refinement (3 iterations) (ablation) | RoomNet basic · LSUN | pixel error (%): 10.46 -> 9.86 | Improved |
| use 2 recurrent iterations (ablation) | RoomNet basic · LSUN | keypoint error (%): 6.95 -> 6.65 | Improved |
| add deep supervision through time to 3-iter model (ablation) | RoomNet recurrent 3-iter · LSUN | keypoint error (%): 6.95 -> 6.30 | Improved |
| add deep supervision through time to 3-iter model (ablation) | RoomNet recurrent 3-iter · LSUN | pixel error (%): 10.47 -> 9.86 | Improved |
| add deep supervision through time to 2-iter model (ablation) | RoomNet recurrent 2-iter · LSUN | keypoint error (%): 6.93 -> 6.65 | Improved |
| add deep supervision through time to 2-iter model (ablation) | RoomNet recurrent 2-iter · LSUN | pixel error (%): 10.44 -> 9.97 | Improved |
| go from 2 to 3 recurrent iterations without deep supervision (ablation) | RoomNet recurrent · LSUN | keypoint error (%): 6.93 -> 6.95 | Hurt |
| stack encoder-decoders instead of single (ablation) | RoomNet basic · LSUN | keypoint error (%): 6.95 -> 6.82 | Improved |
| add skip connections to stacked encoder-decoder (ablation) | Stacked enc-dec · LSUN | keypoint error (%): 6.82 -> 7.05 | Hurt |
| add feedback loop to single encoder-decoder (ablation) | RoomNet basic · LSUN | pixel error (%): 10.46 -> 10.10 | Improved |
| add feedback loop to memory augmented recurrent encoder-decoder (ablation) | RoomNet full · LSUN | keypoint error (%): 6.30 -> 6.37 | No clear effect |
Closest papers by meaning
- LayoutNet: Reconstructing the 3D Room Layout From a Single RGB Image (CVPR 2018 · similarity 0.90)
- PolyLayout: Multi-room Manhattan Layout Estimation (ECCV 2026 · similarity 0.90)
- Structural Deep Metric Learning for Room Layout Estimation (ECCV 2020 · similarity 0.90)
- GeoLayout: Geometry Driven Room Layout Estimation Based on Depth Maps of Planes (ECCV 2020 · similarity 0.90)
- MvFFN: Multi-view Floor-Plan Feed-Forward Network for Unposed Wide-Baseline Panorama Layout Reconstruction (NeurIPS 2026 · similarity 0.88)
- HouseLayout3D: A Benchmark and Training-free Baseline for 3D Layout Estimation in the Wild (NeurIPS 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.
