Research map / 3D vision / Depth and stereo
3D Crowd Counting via Multi-View Fusion with 3D Gaussian Kernels
Read the paperCheck 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 single-view prediction branches with 2D supervision (ablation) | 3D multi-view counting · PETS2009 (7*40cm) | MAE: 4.12 -> 3.20 | Helped |
| add projection consistency measure loss (ablation) | 3D multi-view counting · PETS2009 (7*40cm) | MAE: 3.20 -> 3.15 | Helped |
| add single-view prediction branches with 2D supervision (ablation) | 3D multi-view counting · DukeMTMC (28*10cm) | MAE: 2.15 -> 1.41 | Helped |
| add projection consistency measure loss (ablation) | 3D multi-view counting · DukeMTMC (28*10cm) | MAE: 1.41 -> 1.37 | Helped |
| add single-view prediction branches with 2D supervision (ablation) | 3D multi-view counting · CityStreet (28*10cm) | MAE: 7.87 -> 7.58 | Helped |
| add projection consistency measure loss (ablation) | 3D multi-view counting · CityStreet (28*10cm) | MAE: 7.58 -> 7.54 | Helped |
| increase 3D voxel height resolution from 7*40cm to 28*10cm (ablation) | 3D multi-view counting · PETS2009 | MAE: 3.15 -> 4.21 | Hurt |
| increase 3D voxel height resolution from 7*40cm to 28*10cm (ablation) | 3D multi-view counting · CityStreet | MAE: 8.35 -> 7.54 | Helped |
| replace 2D ground-plane fusion (MVMS) with 3D projection and fusion | multi-view counting · PETS2009 | MAE: 3.49 -> 3.15 | Helped |
| replace 2D ground-plane fusion (MVMS) with 3D projection and fusion | multi-view counting · DukeMTMC | MAE: 1.03 -> 1.37 | No clear effect |
| replace 2D ground-plane fusion (MVMS) with 3D projection and fusion | multi-view counting · CityStreet | MAE: 8.01 -> 7.54 | Helped |
Closest papers by meaning
- Wide-Area Crowd Counting via Ground-Plane Density Maps and Multi-View Fusion CNNs (CVPR 2019 · similarity 0.95)
- Cross-View Cross-Scene Multi-View Crowd Counting (CVPR 2021 · similarity 0.90)
- Single-Image Crowd Counting via Multi-Column Convolutional Neural Network (CVPR 2016 · similarity 0.89)
- Context-Aware Crowd Counting (CVPR 2019 · similarity 0.88)
- Switching Convolutional Neural Network for Crowd Counting (CVPR 2017 · similarity 0.88)
- Crowd Counting via Adversarial Cross-Scale Consistency Pursuit (CVPR 2018 · 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.