Research map / Detection and segmentation / CNN features
Multi-Scale Context Intertwining for Semantic Segmentation
Di Lin, Yuanfeng Ji, Dani Lischinski, Daniel Cohen-Or, Hui Huang
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 bidirectional LSTM context intertwining (RNN) (ablation) | basic FCN · PASCAL VOC 2012 val | mean IoU (%): 77.8 -> 85.1 | Improved |
| add LSTM cell states (ablation) | MSCI · PASCAL VOC 2012 val | mean IoU (%): 84.4 -> 85.1 | Improved |
| add super-pixel guided aggregation (ablation) | MSCI · PASCAL VOC 2012 val | mean IoU (%): 84.3 -> 85.1 | Improved |
| replace SPP with hierarchical bidirectional intertwining | ResNet-152 · PASCAL VOC 2012 val | mean IoU (%): 81.1 -> 85.1 | Improved |
| replace Encoder-Decoder with MSCI | ResNet-152 · PASCAL VOC 2012 val | mean IoU (%): 81.4 -> 85.1 | Improved |
| replace ASPP with MSCI | ResNet-152 · PASCAL VOC 2012 val | mean IoU (%): 82.2 -> 85.1 | Improved |
| replace Encoder-Decoder + ASPP with MSCI | ResNet-152 · PASCAL VOC 2012 val | mean IoU (%): 82.5 -> 85.1 | Improved |
| replace Encoder-Decoder + ASPP with MSCI | ResNet-152 · PASCAL-Context val | mean IoU (%): 47.4 -> 50.3 | Improved |
| add bidirectional connections to existing combination networks (ablation) | SPP, ED, ASPP baselines · PASCAL VOC 2012 val | mean IoU (%): 0.4 ~ 1.2 | Improved |
| use RGB only MSCI instead of RGB-D prior best | NYUDv2 · NYUDv2 val | mean IoU (%): 47.7 -> 49.0 | Improved |
| use RGB only MSCI instead of RGB-D prior best | SUN-RGBD · SUN-RGBD val | mean IoU (%): 48.1 -> 50.4 | Improved |
Closest papers by meaning
- ISNet: Integrate Image-Level and Semantic-Level Context for Semantic Segmentation (ICCV 2021 · similarity 0.91)
- Hierarchically Gated Deep Networks for Semantic Segmentation (CVPR 2016 · similarity 0.91)
- Lightweight and Progressively-Scalable Networks for Semantic Segmentation (IJCV 2023 · similarity 0.89)
- Exploring Context with Deep Structured Models for Semantic Segmentation (TPAMI 2017 · similarity 0.89)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (NeurIPS 2024 · similarity 0.88)
- Multi-Scale Representations by Varying Window Attention for Semantic Segmentation (ICLR 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.
