Research map / Detection and segmentation / CNN features
Regression of Instance Boundary by Aggregated CNN and GCN
Yanda Meng, Wei Meng, Dongxu Gao, Yitian Zhao, Xiaoyun Yang, Xiaowei Huang, Yalin Zheng
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 aggregation module (ablation) | encoder-decoder · fetal head segmentation | Dice: 1.83% | Improved |
| add attention module (ARM) to aggregation (ablation) | aggregation network · fetal head segmentation | Dice: 0.47% | Improved |
| use GCN decoder instead of CNN decoder (ablation) | aggregation + ARM · fetal head segmentation | Dice: 1.11% | Improved |
| add aggregation module (ablation) | encoder-decoder · OD & OC segmentation | Dice: 1.17 % | Improved |
| add attention module (ARM) (ablation) | aggregation network · OD & OC segmentation | Dice: 0.64% | Improved |
| use GCN decoder instead of CNN decoder (ablation) | aggregation + ARM · OD & OC segmentation | Dice: 1.73% | Improved |
| use Fan-loss instead of L1 loss (ablation) | boundary regression · OD & OC segmentation | Dice: 1.6% relatively better | Improved |
| use Fan-loss instead of L1 loss (ablation) | boundary regression · fetal head segmentation | Dice: 2.7% relatively better | Improved |
| add weight mask = 5 to loss (ablation) | Fan-loss · OD & OC segmentation | Dice: 0.79% | Improved |
| vary Fan-loss parameters w and epsilon (ablation) | Fan-loss · OD & OC and FH | Dice | No clear effect |
| add weight mask 7 or 9 instead of 5 (ablation) | Fan-loss · Retina OC segmentation | Dice: 0.9255 -> 0.9175 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- PolarMask — PolarMask: Single Shot Instance Segmentation With Polar Representation (CVPR 2020) · better on 3 of 3 shared results
- DANet — DANet: Divergent Activation for Weakly Supervised Object Localization (ICCV 2019) · better on 3 of 3 shared results
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
- DARCNN: Domain Adaptive Region-Based Convolutional Neural Network for Unsupervised Instance Segmentation in Biomedical Images (CVPR 2021 · similarity 0.87)
- Learning on the Edge: Investigating Boundary Filters in CNNs (IJCV 2019 · similarity 0.85)
- Occlusion-Aware Instance Segmentation Via BiLayer Network Architectures (TPAMI 2023 · similarity 0.84)
- Non-Local U-Nets for Biomedical Image Segmentation (AAAI 2020 · similarity 0.84)
- Container: Context Aggregation Networks (NeurIPS 2021 · similarity 0.84)
- Recurrent Saliency Transformation Network: Incorporating Multi-Stage Visual Cues for Small Organ Segmentation (CVPR 2018 · similarity 0.84)
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.
