Research map / Detection and segmentation / Tracking
Binary Quadratic Programing for Online Tracking of Hundreds of People in Extremely Crowded Scenes
Afshin Dehghan, Mubarak A. Shah
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 motion term to NCC baseline (ablation) | B(NCC) · Seq5 | tracking accuracy (%): 65.34% -> 72.18% | Improved |
| add motion term to NCC baseline (ablation) | B(NCC) · Seq7 | tracking accuracy (%): 66.81% -> 73.80% | Improved |
| add motion term to NCC baseline (ablation) | B(NCC) · Seq1 | tracking accuracy (%): 78.13% -> 76.9% | Hurt |
| add spatial proximity term to B+Mo (ablation) | B+Mo(NCC) · Seq3 | tracking accuracy (%): 87.64% -> 93.70% | Improved |
| add spatial proximity term to B+Mo (ablation) | B+Mo(NCC) · Seq9 | tracking accuracy (%): 64.08% -> 67.73% | Improved |
| add group term to B+Mo (ablation) | B+Mo(NCC) · Seq1 | tracking accuracy (%): 76.9% -> 79.74% | Improved |
| add group term to B+Mo (ablation) | B+Mo(NCC) · Seq9 | tracking accuracy (%): 64.08% -> 64.07% | No clear effect |
| add neighborhood motion term to full NCC model (ablation) | B+Mo+SP+Gr(NCC) · Seq7 | tracking accuracy (%): 74.37% -> 74.0% | Hurt |
| add neighborhood motion term to full NCC model (ablation) | B+Mo+SP+Gr(NCC) · Seq8 | tracking accuracy (%): 93.00% -> 92.13% | Hurt |
| replace NCC template tracker with discriminative KCF (ablation) | B+Mo+SP+Gr+NMo · Seq9 | tracking accuracy (%): 69.80% -> 77.88% | Improved |
| replace NCC with KCF (ablation) | B+Mo+SP+Gr+NMo · Seq7 | tracking accuracy (%): 74.0% -> 67.22% | Hurt |
| use proposed binary QP tracker instead of PNMC | Proposed · Galleria1 | tracking accuracy (%): 86% -> 92% | Improved |
Closest papers by meaning
- Tracking Pedestrian Heads in Dense Crowd (CVPR 2021 · similarity 0.87)
- Fusing Crowd Density Maps and Visual Object Trackers for People Tracking in Crowd Scenes (CVPR 2018 · similarity 0.86)
- Groupwise Tracking of Crowded Similar-Appearance Targets From Low-Continuity Image Sequences (CVPR 2016 · similarity 0.85)
- TAO: A Large-Scale Benchmark for Tracking Any Object (ECCV 2020 · similarity 0.85)
- TrackingNet: A Large-Scale Dataset and Benchmark for Object Tracking in the Wild (ECCV 2018 · similarity 0.85)
- Focusing on Tracks for Online Multi-Object Tracking (CVPR 2025 · 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.
