Research map / Vision-language and multimodal / Cross-modal learning
Multi-modal Frequency Decomposition Network for Semantic Scene Completion
Die Zuo, Lubo Wang, Ruonan Liu, Qing Guo, Chong Wang, Dongdong Wu, Wei Feng, Kairui Yang, Di Lin
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 MAFF to addition-based fusion (ablation) | MFDNet · NYUv2 test | mIoU (%): 54.5 -> 55.7 | Improved |
| add FDC (ablation) | MFDNet · NYUv2 test | mIoU (%): 54.5 -> 55.1 | Improved |
| add both MAFF and FDC (ablation) | MFDNet · NYUv2 test | mIoU (%): 54.5 -> 57.0 | Improved |
| learn fusion weights in frequency domain instead of spatial domain (ablation) | MAFF · NYUv2 test | mIoU (%): 55.5 -> 56.1 | Improved |
| learn weights from overall feature for each band (overall-to-band) instead of band-to-band (ablation) | MAFF · NYUv2 test | mIoU (%): 56.5 -> 57.0 | Improved |
| spatial overall-to-overall weighting versus no weights (ablation) | MAFF · NYUv2 test | IoU (%): 76.3 -> 76.0 | Hurt |
| add only high-frequency components of shallow features instead of overall features (ablation) | FDC · NYUv2 test | mIoU (%): 56.1 -> 56.6 | Improved |
| add learned weights to high-frequency compensation (ablation) | FDC · NYUv2 test | mIoU (%): 56.6 -> 57.0 | Improved |
| learn FDC weights from coarse feature rather than shallow features (ablation) | FDC · NYUv2 test | mIoU (%): 56.3 -> 57.0 | Improved |
| replace AMMNet modulation fusion with MAFF (both without extras) | AMMNet fusion · NYUv2 test | mIoU (%): 55.2 -> 55.7 | Improved |
| use MAFF fusion instead of AMMNet modulation | AMMNet fusion · NYUv2 test | parameters (M): 20.85 -> 4.79 | Improved |
Closest papers by meaning
- FFNet: Frequency Fusion Network for Semantic Scene Completion (AAAI 2022 · similarity 0.92)
- Attention-Based Multi-Modal Fusion Network for Semantic Scene Completion (AAAI 2020 · similarity 0.91)
- Unleashing Network Potentials for Semantic Scene Completion (CVPR 2024 · similarity 0.89)
- Semantic Scene Completion From a Single Depth Image (CVPR 2017 · similarity 0.89)
- Sparsity-Aware Voxel Attention and Foreground Modulation for 3D Semantic Scene Completion (CVPR 2026 · similarity 0.88)
- Anisotropic Convolutional Neural Networks for RGB-D based Semantic Scene Completion (TPAMI 2021 · similarity 0.88)
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.
