Research map / Detection and segmentation / CNN features
Multiview Detection with Feature Perspective Transformation
Yunzhong Hou, Liang Zheng, Stephen Gould
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 large-kernel spatial aggregation (ablation) | MVDet · Wildtrack | MODA: 76.9 -> 88.2 | Improved |
| add large-kernel spatial aggregation (ablation) | MVDet · MultiviewX | MODA: 77.2 -> 83.9 | Improved |
| project feature maps instead of RGB images (ablation) | MVDet · Wildtrack | MODA: 26.8 -> 76.9 | Improved |
| project feature maps instead of RGB images (ablation) | MVDet · MultiviewX | MODA: 19.5 -> 77.2 | Improved |
| project feature maps instead of single-view detection results (ablation) | MVDet · Wildtrack | MODA: 68.2 -> 76.9 | Improved |
| project feature maps instead of single-view detection results (ablation) | MVDet · MultiviewX | MODA: 73.2 -> 77.2 | Improved |
| replace anchor-based DeepMCD with anchor-free feature map projection | MVDet · Wildtrack | MODA: 67.8 -> 76.9 | Improved |
| replace anchor-based DeepMCD with anchor-free feature map projection | MVDet · MultiviewX | MODA: 70.0 -> 77.2 | Improved |
| use large kernel convolution instead of CRF mean-field spatial aggregation (Deep-Occlusion) | MVDet · Wildtrack | MODA: 74.1 -> 88.2 | Improved |
| replace clustering of projected detections with large kernel convolution (ablation) | MVDet · Wildtrack | MODA: 11.3 -> 68.2 | Improved |
| add single view detection loss (alpha=1) (ablation) | MVDet · Wildtrack | MODA: -1.2% without it | Improved |
| add single view detection loss (alpha=1) (ablation) | MVDet · MultiviewX | MODA: -2.0% without it | Improved |
Closest papers by meaning
- Towards Generalizable Multi-Camera 3D Object Detection via Perspective Rendering (AAAI 2025 · similarity 0.86)
- Multi-view Aggregation Network for Dichotomous Image Segmentation (CVPR 2024 · similarity 0.86)
- CaMuViD: Calibration-Free Multi-View Detection (CVPR 2025 · similarity 0.85)
- Multi-View Attentive Contextualization for Multi-View 3D Object Detection (CVPR 2024 · similarity 0.84)
- MVSDet: Multi-View Indoor 3D Object Detection via Efficient Plane Sweeps (NeurIPS 2024 · similarity 0.84)
- Multiview Scene Graph (NeurIPS 2024 · similarity 0.83)
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.
