Research map / Detection and segmentation / CNN features
Instance-Level Segmentation for Autonomous Driving With Deep Densely Connected MRFs
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 |
|---|---|---|---|
| replace the baseline MRF with the full densely connected MRF using DeepLab predictions | DeepLab-LargeFOV instance segmentation · KITTI test set | InsF1: 39.8 -> 56.8 | Helped |
| replace the baseline MRF with the full densely connected MRF using DeepLab predictions | DeepLab-LargeFOV instance segmentation · KITTI test set | MWCov: 68.2 -> 73.7 | Helped |
| add the inter-connected component term (ablation) | LocCNNPred · KITTI test set | InsF1: 28.6 -> 40.4 | Helped |
| add the smoothness term (ablation) | LocCNNPred+InterConnComp · KITTI test set | InsF1: 40.4 -> 51.5 | Helped |
| add post-processing (ablation) | Full (w/ DeepLab) · KITTI test set | InsF1: 56.8 -> 61.1 | Helped |
| use DeepLab-LargeFOV instead of the naive VGG-16 adaptation (ablation) | Full MRF · KITTI test set | InsF1: 51.5 -> 56.8 | Helped |
Closest papers by meaning
- MWSIS: Multimodal Weakly Supervised Instance Segmentation with 2D Box Annotations for Autonomous Driving (AAAI 2024 · similarity 0.89)
- Instance Segmentation by Jointly Optimizing Spatial Embeddings and Clustering Bandwidth (CVPR 2019 · similarity 0.87)
- LWSIS: LiDAR-Guided Weakly Supervised Instance Segmentation for Autonomous Driving (AAAI 2023 · similarity 0.87)
- Weakly Supervised Semantic Segmentation for Driving Scenes (AAAI 2024 · similarity 0.86)
- WISH: Weakly Supervised Instance Segmentation using Heterogeneous Labels (CVPR 2025 · similarity 0.85)
- P2Seg: Pointly-supervised Segmentation via Mutual Distillation (ICLR 2024 · similarity 0.85)
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.