Research map / Learning theory and algorithms / Clustering
SynMVCrowd: A Large Synthetic Benchmark for Multi-view Crowd Counting and Localization
Qi Zhang, Daijie Chen, Yunfei Gong, Hui Huang
Read the paperCheck your idea against 137,217 papers
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 |
|---|---|---|---|
| use ResNet18 backbone instead of VGG19 (ablation) | Baseline (OT) · SynMVCrowd counting | MAE: 59.04 -> 50.20 | Improved |
| use ResNet18 backbone instead of Transformer (ablation) | Baseline (OT) · SynMVCrowd counting | MAE: 66.23 -> 50.20 | Improved |
| use optimal transport loss instead of MSE loss | Baseline · SynMVCrowd multi-view counting | MAE: 68.79 -> 50.20 | Improved |
| use optimal transport loss instead of MSE loss | Baseline · SynMVCrowd multi-view localization | MODA: 34.6 -> 49.6 | Improved |
| add spatial feature selection module (ablation) | Baseline (MSE) · SynMVCrowd multi-view counting | MAE: 74.33 -> 68.79 | Improved |
| add spatial feature selection module (ablation) | Baseline (OT) · SynMVCrowd multi-view counting | MAE: 52.94 -> 50.20 | Improved |
| add spatial feature selection module (ablation) | Baseline (OT) · SynMVCrowd multi-view localization | MODA: 46.7 -> 49.6 | Improved |
| test with 9 camera views instead of 3 views (ablation) | Baseline (OT) · SynMVCrowd multi-view counting | MAE: 57.97 -> 42.46 | Improved |
| test with 9 camera views instead of 3 views (ablation) | CountFormer · SynMVCrowd multi-view counting | MAE: 47.71 -> 77.82 | Hurt |
| use more frontal camera pitch ([-30, 0] instead of [-90,-60)) (ablation) | Baseline (OT) · SynMVCrowd multi-view counting | MAE: 48.19 -> 55.97 | Hurt |
| add domain adaptation to fine-tuning on 100% Wildtrack labels (ablation) | Baseline · Wildtrack | MODA: 85.9 -> 88.7 | Improved |
| add unsupervised domain adaptation to SynMVCrowd-trained counter (ablation) | Baseline counting model · ShanghaiTech A | MAE: 187.70 -> 118.70 | Improved |
Closest papers by meaning
- Cross-View Cross-Scene Multi-View Crowd Counting (CVPR 2021 · similarity 0.90)
- Wide-Area Crowd Counting: Multi-view Fusion Networks for Counting in Large Scenes (IJCV 2022 · similarity 0.90)
- NWPU-Crowd: A Large-Scale Benchmark for Crowd Counting and Localization (TPAMI 2020 · similarity 0.89)
- Rethinking Counting and Localization in Crowds: A Purely Point-Based Framework (ICCV 2021 · similarity 0.89)
- Wide-Area Crowd Counting via Ground-Plane Density Maps and Multi-View Fusion CNNs (CVPR 2019 · similarity 0.89)
- 3D Crowd Counting via Multi-View Fusion with 3D Gaussian Kernels (AAAI 2020 · similarity 0.89)
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.
