Research map / Video / Video segmentation and tracking
Feature Space Optimization for Semantic Video 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 feature space optimization to dense 3D CRF (ablation) | TextonBoost semantic video segmentation · CamVid test | mean IoU: 53.08 -> 55.23 | Helped |
| Add feature space optimization to dense 3D CRF (ablation) | TextonBoost semantic video segmentation · CamVid test | Consistency: 81.68 -> 87.33 | Helped |
| Add dense 3D CRF over the video volume (ablation) | TextonBoost semantic segmentation · CamVid test | mean IoU: 51.08 -> 53.08 | Helped |
| Add dense 3D CRF over the video volume (ablation) | TextonBoost semantic segmentation · CamVid test | Consistency: 74.37 -> 81.68 | Helped |
| Add dense 2D CRF independently within each frame (ablation) | TextonBoost semantic segmentation · CamVid test | mean IoU: 47.43 -> 51.08 | Helped |
| Add dense 2D CRF independently within each frame (ablation) | TextonBoost semantic segmentation · CamVid test | Consistency: 60.88 -> 74.37 | Helped |
| Use Discrete Flow instead of LDOF for feature space optimization (ablation) | Dilation semantic video segmentation · CamVid test | mean IoU: 66.11 -> 66.12 | No clear effect |
| Use feature space optimization and spatio-temporal CRF | Dilation · CamVid test | mean IoU: 65.29 -> 66.11 | Helped |
| Use feature space optimization and spatio-temporal CRF | Dilation · CamVid test | Consistency: 79.0 -> 88.3 | Helped |
| Use feature space optimization and spatio-temporal CRF | Dilation unary · Cityscapes validation | mean IoU: 68.65 -> 70.30 | Helped |
| Use feature space optimization and spatio-temporal CRF | Dilation unary · Cityscapes validation | Consistency: 88.14 -> 94.71 | Helped |
Closest papers by meaning
- Low-Latency Video Semantic Segmentation (CVPR 2018 · similarity 0.89)
- Learning Spatial-Semantic Features for Robust Video Object Segmentation (ICLR 2025 · similarity 0.89)
- Exploiting Temporal State Space Sharing for Video Semantic Segmentation (CVPR 2025 · similarity 0.89)
- Every Frame Counts: Joint Learning of Video Segmentation and Optical Flow (AAAI 2020 · similarity 0.89)
- Exploring Temporal Feature Correlation for Efficient and Stable Video Semantic Segmentation (AAAI 2024 · similarity 0.88)
- Re-thinking Temporal Search for Long-Form Video Understanding (CVPR 2025 · 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.