Research map / 3D vision / Depth and stereo
BEV-Net: Assessing Social Distancing Compliance by Joint People Localization and Geometric Reasoning
Zhirui Dai, Yuepeng Jiang, Yi Li, Bo Liu, Antoni B. Chan, Nuno Vasconcelos
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 |
|---|---|---|---|
| add feet branch to head-only model (ablation) | BEV-Net · CityUHK-X-BEV | local risk map IoU (%): 67.43 -> 71.25 | Improved |
| add feet branch to head-only model (ablation) | BEV-Net · CityUHK-X-BEV | BEV heatmap MSE x1e-7: 2.03 -> 1.34 | Improved |
| add feet branch to head-only model (ablation) | BEV-Net · CityUHK-X-BEV | global risk MSE x1e-4: 8.95 -> 6.24 | Improved |
| add head branch to feet-only model (ablation) | BEV-Net · CityUHK-X-BEV | local risk map IoU (%): 68.12 -> 71.25 | Improved |
| add head branch to feet-only model (ablation) | BEV-Net · CityUHK-X-BEV | global risk MSE x1e-4: 7.62 -> 6.24 | Improved |
| add head branch to feet-only model (ablation) | BEV-Net · CityUHK-X-BEV | BEV heatmap MSE x1e-7: 1.38 -> 1.34 | Improved |
| add group BEV-Transform with spatial attention to head branch (ablation) | BEV-Net · CityUHK-X-BEV | local risk map IoU (%): 69.65 -> 71.25 | Improved |
| add group BEV-Transform with spatial attention to head branch (ablation) | BEV-Net · CityUHK-X-BEV | global risk MSE x1e-4: 7.08 -> 6.24 | Improved |
| add group BEV-Transform with spatial attention to head branch (ablation) | BEV-Net · CityUHK-X-BEV | normalized Chamfer distance Dn (%): 41.3 -> 41.7 | Hurt |
| add group BEV-Transform with spatial attention to head branch (ablation) | BEV-Net · CityUHK-X-BEV | BEV heatmap MSE x1e-7: 1.36 -> 1.34 | Improved |
| use BEV-Net instead of ground-truth-head crowd-counting oracle | CC oracle · CityUHK-X-BEV | global risk MSE x1e-4: 48.13 -> 6.24 | Improved |
| use BEV-Net instead of Faster R-CNN detection baseline | Faster R-CNN · CityUHK-X-BEV | local risk map IoU (%): 67.01 -> 71.25 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CSRNet — CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes (CVPR 2018) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Human Mobility Modeling during the COVID-19 Pandemic via Deep Graph Diffusion Infomax (AAAI 2023 · similarity 0.81)
- MINGLE: VLMs for Semantically Complex Region Detection in Urban Scenes (AAAI 2026 · similarity 0.81)
- Recurrent Attentive Zooming for Joint Crowd Counting and Precise Localization (CVPR 2019 · similarity 0.80)
- Characterizing and Discovering Spatiotemporal Social Contact Patterns for Healthcare (TPAMI 2016 · similarity 0.80)
- Map It Anywhere: Empowering BEV Map Prediction using Large-scale Public Datasets (NeurIPS 2024 · similarity 0.79)
- Promoting Fair Vaccination Strategies through Influence Maximization: A Case Study on COVID-19 Spread (AAAI 2024 · similarity 0.78)
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.
