Research map / Video / Video-language understanding
WebUAV-3M: A Benchmark for Unveiling the Power of Million-Scale Deep UAV Tracking
Chunhui Zhang, Guanjie Huang, Li Liu, Shan Huang, Yinan Yang, Xiang Wan, Shiming Ge, Dacheng Tao
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 SATA instead of CVAT for annotation | semi-automatic annotation · annotation time | mean time per box: 3.86 s -> 2.99 s | Improved |
| use SATA instead of manual Labelme | annotation · annotation time | mean time per box: 15.81 s -> 2.99 s | Improved |
| add a refinement module (AlphaRefine) to DiMP | DiMP · WebUAV-3M test set | AUC: 0.364 -> 0.593 | Improved |
| use probabilistic regression (PrDiMP) in place of DiMP | DiMP · WebUAV-3M test set | AUC: 0.364 -> 0.514 | Improved |
| use handcrafted-feature MCCT versus deep ATOM | tracker choice · WebUAV-3M test set | AUC: 0.291 -> 0.440 | Improved |
| classic ECO versus recent ATOM | tracker choice · WebUAV-3M test set | AUC: 0.291 -> 0.450 | Improved |
| AlphaRefine versus second-best tracker | tracker choice · WebUAV-3M test set | Pre, nPre, AUC, cAUC, mAcc: 4.3%, 4.0%, 5.0%, 5.0% and 5.2% | Improved |
| lightweight AlexNet-based SiamRPN vs deeper trackers | speed · WebUAV-3M test set | FPS: 142.8@GPU vs 32.4@GPU (PrDiMP) | Improved |
| retrain five trackers on WebUAV-3M training set | SiamFC, SiamRPN, ATOM, GOTURN, MDNet · WebUAV-3M test set | Pre, nPre, AUC, cAUC, mAcc | Improved |
| increase adversarial perturbation iterations | most trackers · adversarial examples subtest | mAcc | No clear effect |
| add adversarial examples to input | TransT and MDNet · adversarial examples subtest | mAcc | Hurt |
| decrease target size below 30 | all trackers · small targets subtest | mAcc | Hurt |
Closest papers by meaning
- UAV3D: A Large-scale 3D Perception Benchmark for Unmanned Aerial Vehicles (NeurIPS 2024 · similarity 0.89)
- MUST: The First Dataset and Unified Framework for Multispectral UAV Single Object Tracking (CVPR 2025 · similarity 0.88)
- JTD-UAV: MLLM-Enhanced Joint Tracking and Description Framework for Anti-UAV Systems (CVPR 2025 · similarity 0.87)
- WebUOT-1M: Advancing Deep Underwater Object Tracking with A Million-Scale Benchmark (NeurIPS 2024 · similarity 0.86)
- Dynamic Semantic-Aware Correlation Modeling for UAV Tracking (NeurIPS 2025 · similarity 0.86)
- MMOT: The First Challenging Benchmark for Drone-based Multispectral Multi-Object Tracking (NeurIPS 2025 · similarity 0.86)
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.
