Research map / Detection and segmentation / Tracking
AerialMind: Towards Referring Multi-Object Tracking in UAV Scenarios
Chenglizhao Chen, Shaofeng Liang, Runwei Guan, Xiaolou Sun, Haocheng Zhao, Haiyun Jiang, Tao Huang, Henghui Ding, Qing-Long Han
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 CFE and SACR (ablation) | HETrack · AerialMind in-domain | HOTA: 26.41 -> 31.46 | Improved |
| add co-evolutionary fusion encoder (CFE) (ablation) | HETrack · AerialMind in-domain | HOTA: 28.27 -> 31.46 | Improved |
| add scale adaptive contextual refinement (SACR) (ablation) | HETrack · AerialMind in-domain | HOTA: 29.89 -> 31.46 | Improved |
| use full semantic guidance module instead of concatenation (ablation) | HETrack · AerialMind in-domain | HOTA: 28.88 -> 31.46 | Improved |
| use full semantic guidance instead of cross-attention with Ts (ablation) | HETrack · AerialMind in-domain | HOTA: 30.52 -> 31.46 | Improved |
| use full semantic guidance instead of cross-attention with Ts (ablation) | HETrack · AerialMind in-domain | AssA: 48.82 -> 46.23 | Hurt |
| use full semantic guidance instead of addition (ablation) | HETrack · AerialMind in-domain | AssA: 46.65 -> 46.23 | Hurt |
| HETrack vs best prior method on in-domain test | HETrack vs MGLT · AerialMind in-domain | HOTA: 26.16 -> 31.46 | Improved |
| HETrack vs TempRMOT on association | HETrack vs TempRMOT · AerialMind in-domain | AssA: 53.22 -> 46.23 | Hurt |
| HETrack vs TransRMOT on localization | HETrack vs TransRMOT · AerialMind in-domain | LocA: 83.47 -> 82.77 | Hurt |
| evaluate HETrack on cross-domain UAVDT | HETrack · AerialMind cross-domain (UAVDT) | HOTA: 27.58 -> 31.60 | Improved |
| HETrack on ground-level Refer-KITTI-V2 | HETrack · Refer-KITTI-V2 | HOTA: 36.18 vs 35.40 | No clear effect |
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.
- MOTRv2 — MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors (CVPR 2023) · 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
- AirCopBench: A Benchmark for Multi-drone Collaborative Embodied Perception and Reasoning (AAAI 2026 · similarity 0.86)
- Towards Realistic UAV Vision-Language Navigation: Platform, Benchmark, and Methodology (ICLR 2025 · similarity 0.85)
- UAV-Flow Colosseo: A Real-World Benchmark for Flying-on-a-Word UAV Imitation Learning (NeurIPS 2025 · similarity 0.85)
- JTD-UAV: MLLM-Enhanced Joint Tracking and Description Framework for Anti-UAV Systems (CVPR 2025 · similarity 0.85)
- Dynamic Semantic-Aware Correlation Modeling for UAV Tracking (NeurIPS 2025 · similarity 0.85)
- MMOT: The First Challenging Benchmark for Drone-based Multispectral Multi-Object Tracking (NeurIPS 2025 · similarity 0.85)
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.
