Research map / Detection and segmentation / CNN features
MIML-FCN+: Multi-Instance Multi-Label Learning via Fully Convolutional Networks With Privileged Information
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 bounding boxes as privileged information | MIML-FCN · VOC 2007 | mAP: 90.2 -> 92.4 | Helped |
| Add bounding boxes as privileged information | MIML-FCN · VOC 2012 | mAP: 89.8 -> 91.9 | Helped |
| Add bounding boxes as privileged information | MIML-FCN · MS COCO | mAP: 63.5 -> 65.6 | Helped |
| Add image captions as privileged information | MIML-FCN · MS COCO | mAP: 63.5 -> 64.6 | Helped |
| Exploit nearest-neighbor inter-instance correlations | MIML-FCN+BB · VOC 2007 | mAP: 92.4 -> 93.1 | Helped |
| Exploit nearest-neighbor inter-instance correlations | MIML-FCN+BB · VOC 2012 | mAP: 91.9 -> 92.5 | Helped |
| Exploit nearest-neighbor inter-instance correlations | MIML-FCN+BB · MS COCO | mAP: 65.6 -> 66.2 | Helped |
| Exploit nearest-neighbor inter-instance correlations | MIML-FCN+CP · MS COCO | mAP: 64.6 -> 65.4 | Helped |
| Add a nonlinear convolutional-ReLU feature-mapping layer (ablation) | MIML-FCN · VOC 2007 | accuracy | Helped |
| Use nonlinear rather than linear feature mapping | MIML-FCN · VOC 2007 | accuracy: more than 2% | Helped |
Closest papers by meaning
- Deep MIML Network (AAAI 2017 · similarity 0.89)
- Discover Multiple Novel Labels in Multi-Instance Multi-Label Learning (AAAI 2017 · similarity 0.89)
- Scalable Optimization of Multivariate Performance Measures in Multi-instance Multi-label Learning (AAAI 2017 · similarity 0.87)
- T-MDML: Triplet-based Multiple Distance Metric Learning for Multi-Instance Multi-Label Classification with Label Correlation (AAAI 2025 · similarity 0.84)
- Multi-Instance Partial-Label Learning with Margin Adjustment (NeurIPS 2024 · similarity 0.84)
- Taming the Sigmoid Bottleneck: Provably Argmaxable Sparse Multi-Label Classification (AAAI 2024 · 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.