Research map / Vision-language and multimodal / Cross-modal learning
Mix and Match Networks: Cross-Modal Alignment for Zero-Pair Image-to-Image Translation
Yaxing Wang, Luis Herranz, Joost van de Weijer
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 |
|---|---|---|---|
| use skip connections instead of no side information (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 29.8% -> 12.7% | Hurt |
| use pooling indices as side information (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 29.8% -> 43.2% | Improved |
| initialize RGB encoder with ImageNet pretrained weights (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 43.2% -> 46.7% | Improved |
| add autoencoders (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 6.48% -> 20.3% | Improved |
| add latent space consistency loss (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 20.3% -> 45.8% | Improved |
| add Gaussian noise to encoder output (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 45.8% -> 46.7% | Improved |
| add pseudo-pairs (ablation) | M&MNets · SceneNet RGB-D (51K subset) D->S | mIoU: 46.7% -> 49.2% | Improved |
| add pseudo-pairs (ablation) | M&MNets · SceneNet RGB-D full D->S | mIoU: 55.4 -> 60.1 | Improved |
| add pseudo-pairs (ablation) | M&MNets · SUN RGB-D D->S | mIoU: 28.1 -> 30.2 | Improved |
| add pseudo-pairs (ablation) | M&MNets · Flowers Theta2->Theta3 zero-pair | accuracy: 36.5 -> 57.5 | Improved |
| add RGB to depth input (multi-modal fusion) | M&MNets · SceneNet RGB-D to segmentation | mIoU: 55.4 -> 57.1 | Improved |
| add skip connections when all translations are seen (D->R->S) | M&MNets · SceneNet RGB-D D->R->S | mIoU: 5.96 -> 24.7 | Improved |
Closest papers by meaning
- Do Vision and Language Encoders Represent the World Similarly? (CVPR 2024 · similarity 0.88)
- Linguistic-Aware Patch Slimming Framework for Fine-grained Cross-Modal Alignment (CVPR 2024 · similarity 0.88)
- SoftCLIP: Softer Cross-Modal Alignment Makes CLIP Stronger (AAAI 2024 · similarity 0.88)
- Dynamic Adapter with Semantics Disentangling for Cross-lingual Cross-modal Retrieval (AAAI 2025 · similarity 0.88)
- CrossOver: 3D Scene Cross-Modal Alignment (CVPR 2025 · similarity 0.88)
- CL2CM: Improving Cross-Lingual Cross-Modal Retrieval via Cross-Lingual Knowledge Transfer (AAAI 2024 · 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.
