Research map / Label-efficient and robust learning / Domain adaptation and generalization
Learning From Synthetic Data for Crowd Counting in the Wild
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 |
|---|---|---|---|
| Pretrain on GCC, then fine-tune on real labeled data (ablation) | MCNN · UCF-QNRF | MAE: 281.2 -> 199.8 | Helped |
| Pretrain on GCC, then fine-tune on real labeled data (ablation) | SFCN · UCF-QNRF | MAE: 134.3 -> 124.7 | Helped |
| Use ResNet-101 instead of VGG-16 as the SFCN backbone, with GCC pretraining | SFCN · UCF-QNRF | MAE: 124.7 -> 102.0 | Helped |
| Use GCC pretraining and fine-tuning for crowd counting | SFCN† · UCF-QNRF | MAE: 132 -> 102.0 | Helped |
| Use GCC pretraining and fine-tuning for crowd counting | SFCN† · SHT A | MAE: 67.0 -> 64.8 | Helped |
| Use GCC pretraining and fine-tuning for crowd counting | SFCN† · SHT B | MAE: 8.4 -> 7.6 | Helped |
| Use GCC pretraining and fine-tuning for crowd counting | SFCN† · UCF_CC_50 | MAE: 219.2 -> 214.2 | Helped |
| Use GCC pretraining and fine-tuning for crowd counting | SFCN† · WorldExpo'10 | MAE: 7.5 -> 9.4 | Hurt |
| Add SSIM embedding to Cycle GAN (ablation) | Crowd-counting domain adaptation · SHT A | MAE: 143.3 -> 123.4 | Helped |
| Add SSIM embedding to Cycle GAN (ablation) | Crowd-counting domain adaptation · SHT B | MAE: 25.4 -> 19.9 | Helped |
| Add SSIM embedding to Cycle GAN (ablation) | Crowd-counting domain adaptation · UCF_CC_50 | MAE: 404.6 -> 373.4 | Helped |
| Add SSIM embedding to Cycle GAN (ablation) | Crowd-counting domain adaptation · WorldExpo'10 | Avg. MAE: 32.4 -> 26.3 | Helped |
Closest papers by meaning
- Leveraging Self-Supervision for Cross-Domain Crowd Counting (CVPR 2022 · similarity 0.91)
- Adapting Lightweight Image-based Counting Models for Video Crowd Counting (CVPR 2026 · similarity 0.88)
- Domain-General Crowd Counting in Unseen Scenarios (AAAI 2023 · similarity 0.88)
- CrowdDiff: Multi-hypothesis Crowd Density Estimation using Diffusion Models (CVPR 2024 · similarity 0.86)
- Taste More, Taste Better: Diverse Data and Strong Model Boost Semi-Supervised Crowd Counting (CVPR 2025 · similarity 0.86)
- Single Domain Generalization for Crowd Counting (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.