Research map / Detection and segmentation / Open-set and few-shot detection
Segmenting Transparent Objects in the Wild
Enze Xie, Wenjia Wang, Wenhai Wang, Mingyu Ding, Chunhua Shen, Ping Luo
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 boundary attention module and boundary stream | DeepLabV3+ · Trans10K test (Hard) | mIoU (%): 69.0 -> 72.1 | Improved |
| add Dice boundary loss as auxiliary loss (ablation) | DeepLabV3+ baseline · Trans10K (Hard) | mIoU (%): +1.25% | Improved |
| add Dice boundary loss as auxiliary loss (ablation) | DeepLabV3+ baseline · Trans10K (Hard) | Acc (%): +2.31% | Improved |
| use any boundary auxiliary loss (BCE, Focal or Dice) (ablation) | DeepLabV3+ baseline · Trans10K (Hard) | mIoU (%) | Improved |
| use Focal loss instead of BCE for the boundary (ablation) | boundary stream · Trans10K (Hard) | mIoU (%) | Improved |
| use Dice loss instead of Focal loss for the boundary (ablation) | boundary stream · Trans10K (Hard) | mIoU (%) | Improved |
| apply boundary attention at all three feature levels instead of fewer (ablation) | TransLab · Trans10K (Hard) | mIoU (%): 70.29 -> 72.10 | Improved |
| apply boundary attention at all three feature levels instead of fewer (ablation) | TransLab · Trans10K (Hard) | Acc (%): 80.38 -> 83.04 | Improved |
| use TransLab instead of DeepLabV3+ | DeepLabV3+ · Trans10K test (Hard) | Acc (%): +3.97% | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- BiSeNet — BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation (ECCV 2018) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Enhanced Boundary Learning for Glass-Like Object Segmentation (ICCV 2021 · similarity 0.88)
- Learning Transparent Object Matting (IJCV 2019 · similarity 0.85)
- TransNormal: Dense Visual Semantics for Diffusion-based Transparent Object Normal Estimation (ICML 2026 · similarity 0.84)
- Glass Segmentation Using Intensity and Spectral Polarization Cues (CVPR 2022 · similarity 0.84)
- High-Precision Dichotomous Image Segmentation via Probing Diffusion Capacity (ICLR 2025 · similarity 0.83)
- unMORE: Unsupervised Multi-Object Segmentation via Center-Boundary Reasoning (ICML 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.
