Research map / Detection and segmentation / CNN features
Fully Convolutional Instance-Aware Semantic Segmentation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 position-sensitive score maps instead of translation-invariant maps (ablation) | FCIS · PASCAL VOC 2012 val | mAP^r@0.5 (%): 52.5 -> 65.7 | Helped |
| add position-sensitive score maps instead of translation-invariant maps (ablation) | FCIS · PASCAL VOC 2012 val | mAP^r@0.7 (%): 38.5 -> 52.1 | Helped |
| jointly formulate mask prediction and classification rather than use separate score maps (ablation) | FCIS · PASCAL VOC 2012 val | mAP^r@0.5 (%): 63.9 -> 65.7 | Helped |
| jointly formulate mask prediction and classification rather than use separate score maps (ablation) | FCIS · PASCAL VOC 2012 val | mAP^r@0.7 (%): 49.7 -> 52.1 | Helped |
| replace naïve MNC's per-ROI fully connected mask head with FCIS | instance-aware segmentation · PASCAL VOC 2012 val | mAP^r@0.7 (%): 36.0 -> 52.1 | Helped |
| replace MNC with FCIS | instance-aware segmentation · COCO test-dev | mAP^r@[0.5:0.95] (%): 24.6 -> 28.8 | Helped |
| replace MNC with FCIS | instance-aware segmentation · COCO test-dev | test time/img: 1.37s -> 0.24s | Helped |
| use OHEM | FCIS · COCO test-dev | mAP^r@[0.5:0.95] (%): 28.8 -> 29.2 | Helped |
| increase backbone depth from ResNet-50 to ResNet-101 (ablation) | FCIS · COCO test-dev | mAP^r@[0.5:0.95] (%): 27.1 -> 29.2 | Helped |
| increase backbone depth from ResNet-101 to ResNet-152 (ablation) | FCIS · COCO test-dev | mAP^r@[0.5:0.95] (%): 29.2 -> 29.5 | Helped |
| add multi-scale testing (ablation) | FCIS · COCO test-dev | mAP^r@[0.5:0.95] (%): 29.2 -> 32.0 | Helped |
| add horizontal-flip testing after multi-scale testing (ablation) | FCIS · COCO test-dev | mAP^r@[0.5:0.95] (%): 32.0 -> 32.7 | Helped |
Closest papers by meaning
- Instance-Aware Semantic Segmentation via Multi-Task Network Cascades (CVPR 2016 · similarity 0.92)
- SIM: Semantic-Aware Instance Mask Generation for Box-Supervised Instance Segmentation (CVPR 2023 · similarity 0.90)
- BoxInst: High-Performance Instance Segmentation With Box Annotations (CVPR 2021 · similarity 0.90)
- Boundary-Aware Instance Segmentation (CVPR 2017 · similarity 0.89)
- Deep Variational Instance Segmentation (NeurIPS 2020 · similarity 0.89)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (CVPR 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.