Research map / 3D vision / 3D generation
Projecting Your View Attentively: Monocular Road Scene Layout Estimation via Cross-View Transformation
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 MLP view projection to the vanilla encoder-decoder (ablation) | Proposed framework · KITTI 3D Object | mIOU (%): 22.31 -> 27.42 | Helped |
| Add cross-view correlation attention (ablation) | Proposed framework · KITTI 3D Object | mIOU (%): 27.42 -> 35.03 | Helped |
| Add cycle structure and cycle loss (ablation) | Proposed framework · KITTI 3D Object | mIOU (%): 35.03 -> 35.54 | Helped |
| Add feature selection to the cross-view transformer (ablation) | Proposed framework · KITTI 3D Object | mIOU (%): 35.54 -> 38.79 | Helped |
| Use original features X as key and cycled features X′′ as value in CVT (ablation) | Cross-view transformer · KITTI 3D Object | mIOU (%): 37.58 -> 38.79 | Helped |
| Use projected features X′ as both key and value instead of X and X′′ (ablation) | Cross-view transformer · KITTI 3D Object | mIOU (%): 38.79 -> 25.43 | Hurt |
| Use cycled features X′′ as both key and value instead of X and X′′ (ablation) | Cross-view transformer · KITTI 3D Object | mIOU (%): 38.79 -> 27.71 | Hurt |
| Use original features X as both key and value instead of X and X′′ (ablation) | Cross-view transformer · KITTI 3D Object | mIOU (%): 38.79 -> 35.13 | Hurt |
| Replace the context-aware discriminator with standard PatchGAN (ablation) | Proposed generator · Argoverse Vehicle | mIOU (%): 48.48 -> 47.07 | Hurt |
| Replace the context-aware discriminator with standard PatchGAN (ablation) | Proposed generator · Argoverse Vehicle | mAP (%): 64.04 -> 62.33 | Hurt |
| Add the context-aware discriminator instead of using no discriminator (ablation) | Proposed generator · Argoverse Vehicle | mIOU (%): 48.10 -> 48.48 | Helped |
| Add the context-aware discriminator instead of using no discriminator (ablation) | Proposed generator · Argoverse Vehicle | mAP (%): 61.94 -> 64.04 | Helped |
Closest papers by meaning
- Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis (NeurIPS 2026 · similarity 0.87)
- SDGOCC: Semantic and Depth-Guided Bird's-Eye View Transformation for 3D Multimodal Occupancy Prediction (CVPR 2025 · similarity 0.87)
- StreetCrafter: Street View Synthesis with Controllable Video Diffusion Models (CVPR 2025 · similarity 0.86)
- LidaRF: Delving into Lidar for Neural Radiance Field on Street Scenes (CVPR 2024 · similarity 0.86)
- Unveiling the Hidden: Online Vectorized HD Map Construction with Clip-Level Token Interaction and Propagation (NeurIPS 2024 · similarity 0.85)
- MagicDrive: Street View Generation with Diverse 3D Geometry Control (ICLR 2024 · similarity 0.85)
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.