Research map / Image generation / Image restoration
Disentangled Image Matting
Shaofan Cai, Xiaoshuai Zhang, Haoqiang Fan, Haibin Huang, Jiangyu Liu, Jiaming Liu, Jiaying Liu, Jue Wang, Jian Sun
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 the propagation unit (ablation) | AdaMatting · Matting module ablation | SAD: 44.13 -> 41.70 | Improved |
| add the propagation unit (ablation) | AdaMatting · Matting module ablation | Grad: 17.86 -> 16.89 | Improved |
| add the propagation unit (ablation) | AdaMatting · Matting module ablation | MSE: 0.0111 -> 0.0102 | Improved |
| add the trimap-adaptation decoder (ablation) | AdaMatting · Matting module ablation | SAD: 46.68 -> 41.70 | Improved |
| add the trimap-adaptation decoder (ablation) | AdaMatting · Matting module ablation | MSE: 0.0129 -> 0.0102 | Improved |
| add global convolutions (given no sub-pixel and no propagation unit) (ablation) | AdaMatting · Matting module ablation | SAD: 51.45 -> 45.71 | Improved |
| add sub-pixel convolutions (given no global convolutions and no propagation unit) (ablation) | AdaMatting · Matting module ablation | SAD: 51.45 -> 47.32 | Improved |
| add sub-pixel and global convolutions (given no propagation unit) (ablation) | AdaMatting · Matting module ablation | SAD: 51.45 -> 44.13 | Improved |
| use dynamically weighted multi-task loss instead of fixed weights (ablation) | AdaMatting · Multi-task loss analysis | SAD: 43.54 -> 41.70 | Improved |
| use dynamically weighted multi-task loss instead of fixed weights (ablation) | AdaMatting · Multi-task loss analysis | Grad: 18.35 -> 16.89 | Improved |
| use dynamically weighted multi-task loss instead of fixed weights (ablation) | AdaMatting · Multi-task loss analysis | MSE: 0.0118 -> 0.0102 | Improved |
Closest papers by meaning
- Semantic Image Matting: General and Specific Semantics (IJCV 2023 · similarity 0.87)
- Image Matting and 3D Reconstruction in One Loop (IJCV 2025 · similarity 0.87)
- Natural Image Matting Using Deep Convolutional Neural Networks (ECCV 2016 · similarity 0.85)
- Deep Automatic Portrait Matting (ECCV 2016 · similarity 0.85)
- Training Matting Models Without Alpha Labels (AAAI 2025 · similarity 0.83)
- MP-Mat: A 3D-and-Instance-Aware Human Matting and Editing Framework with Multiplane Representation (ICLR 2025 · similarity 0.83)
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.
