Research map / Point clouds / Segmentation and detection
Focus for Free in Density-Based Counting
Zenglin Shi, Pascal Mettes, Cees G. M. Snoek
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 occlusion simulation | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 60.1 -> 57.2 | Improved |
| add occlusion simulation | T2T-ViT-14 · UCF_QNRF | MAE: 90.8 -> 87.1 | Improved |
| occlusion simulation instead of CutMix | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 58.6 -> 57.2 | Improved |
| add occlusion simulation (high occlusion subset) | T2T-ViT-14 · ShanghaiTech Part_A high-occlusion | MAE: 94.7 -> 91.7 | Improved |
| add foreground distillation | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 60.1 -> 56.5 | Improved |
| add foreground distillation | T2T-ViT-14 · UCF_QNRF | MAE: 90.8 -> 86.7 | Improved |
| standard distillation | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 60.1 -> 59.8 | No clear effect |
| add local segmentation focus (ablation) | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 60.1 -> 58.4 | Improved |
| add global density focus (ablation) | T2T-ViT-14 · UCF_QNRF | MAE: 90.8 -> 87.6 | Improved |
| combine local and global focus (ablation) | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 58.4 -> 56.9 | Improved |
| add focus for free to occlusion simulation + foreground distillation (ablation) | T2T-ViT-14 · ShanghaiTech Part_A | MAE: 55.8 -> 51.6 | Improved |
| add occlusion simulation to foreground distillation + focus for free (ablation) | T2T-ViT-14 · UCF_QNRF | MAE: 82.6 -> 80.9 | Improved |
Closest papers by meaning
- Counting With Focus for Free (ICCV 2019 · similarity 0.93)
- Kernel-Based Density Map Generation for Dense Object Counting (TPAMI 2020 · similarity 0.90)
- Active Crowd Counting with Limited Supervision (ECCV 2020 · similarity 0.90)
- Point-to-Region Loss for Semi-Supervised Point-Based Crowd Counting (CVPR 2025 · similarity 0.89)
- Learning to Count without Annotations (CVPR 2024 · similarity 0.88)
- A Fixed-Point Approach to Unified Prompt-Based Counting (AAAI 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.
