Research map / Detection and segmentation / CNN features
End-To-End Instance Segmentation With Recurrent Attention
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 recurrent attention with an external memory and instance-wise segmentation | instance segmentation · CVPPP A1 test | SBD: 74.4 (4.3) -> 84.9 (4.8) | Helped |
| add recurrent attention with an external memory and instance-wise segmentation | instance segmentation · KITTI test | MUCov: 75.8 -> 66.9 | Hurt |
| add recurrent attention with an external memory and instance-wise segmentation | instance segmentation · Cityscapes all classes | AP: 8.9 -> 9.5 | Helped |
| add recurrent attention with an external memory and instance-wise segmentation | instance segmentation · Cityscapes all classes | AP50%: 21.1 -> 18.9 | Hurt |
| add recurrent attention with an external memory and instance-wise segmentation | instance segmentation · Cityscapes car | AP: 22.5 -> 27.5 | Helped |
| add foreground and angle-map preprocessing (ablation) | recurrent attention model · KITTI validation | MWCov: 55.6 -> 75.1 | Helped |
| add the box network to localize objects before segmentation (ablation) | recurrent attention model · KITTI validation | MUCov: 49.1 -> 64.6 | Helped |
| add angle-map prediction to foreground preprocessing (ablation) | recurrent attention model · KITTI validation | MWCov: 71.2 -> 75.1 | Helped |
| add scheduled sampling so training gradually uses model predictions (ablation) | recurrent attention model · KITTI validation | MUCov: 63.9 -> 64.6 | Helped |
| increase box-network glimpses from one to five (ablation) | recurrent attention model · KITTI validation | MWCov: 64.1 -> 75.1 | Helped |
| increase box-network glimpses from three to five (ablation) | recurrent attention model · KITTI validation | MUCov: 63.4 -> 64.6 | Helped |
Closest papers by meaning
- Fully Convolutional Instance-Aware Semantic Segmentation (CVPR 2017 · similarity 0.87)
- Instance-Aware Semantic Segmentation via Multi-Task Network Cascades (CVPR 2016 · similarity 0.87)
- Recurrent Scene Parsing With Perspective Understanding in the Loop (CVPR 2018 · similarity 0.87)
- Referring Image Segmentation via Recurrent Refinement Networks (CVPR 2018 · similarity 0.87)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (CVPR 2024 · similarity 0.86)
- CLIP as RNN: Segment Countless Visual Concepts without Training Endeavor (CVPR 2024 · similarity 0.86)
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.