Research map / Video / Video segmentation and tracking
Watching Swarm Dynamics from Above: A Framework for Advanced Object Tracking in Drone Videos
Pia Bideau, Duc Pham, Félicie Dhellemmes, Matthew J. Hansen, Jens Krause
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 particle filter tracking on top of per-frame DeepLabV3 segmentation | DeepLabV3 · Drone fish-school tracking (400 samples) | SDR S=30px: 69.31 -> 84.40 | Improved |
| add particle filter tracking on top of DeepLabV3 with only 16 training samples | DeepLabV3 · Drone fish-school tracking (16 samples) | SDR S=30px: 31.63 -> 76.66 | Improved |
| reduce training samples from 400 to 16 | SwDA · Drone fish-school tracking | SDR S=30px: 84.40 -> 76.66 | Hurt |
| reduce training samples from 400 to 16 | DeepLabV3 · Drone fish-school tracking | SDR S=30px: 69.31 -> 31.63 | Hurt |
| reduce training samples from 400 to 16 | Follow Anything · Drone fish-school tracking | SDR S=30px: 37.96 -> 38.37 | No clear effect |
| add particle filter on DeepLabV3 masks, shape via alpha-shape | DeepLabV3 · Fish-school shape segmentation | IoU / Recall: IoU 73.8 -> 71.4; recall 77.4 -> 85.4 | Mixed |
| fuse IMU and GPS with Kalman filter instead of GPS alone (ablation) | GPS · 3D marker localization | absolute error [m]: 0.41 -> 0.32 | Improved |
| fuse IMU and GPS with Kalman filter instead of IMU alone (ablation) | IMU · 3D marker localization | absolute error [m]: 0.96 -> 0.32 | Improved |
Closest papers by meaning
- Towards Dense Object Tracking in a 2D Honeybee Hive (CVPR 2018 · similarity 0.87)
- BirdCollect: A Comprehensive Benchmark for Analyzing Dense Bird Flock Attributes (AAAI 2024 · similarity 0.85)
- JRDB-PanoTrack: An Open-world Panoptic Segmentation and Tracking Robotic Dataset in Crowded Human Environments (CVPR 2024 · similarity 0.85)
- NetTrack: Tracking Highly Dynamic Objects with a Net (CVPR 2024 · similarity 0.84)
- BuckTales: A multi-UAV dataset for multi-object tracking and re-identification of wild antelopes (NeurIPS 2024 · similarity 0.84)
- GMOS: Grounding Moving Object Segmentation in 3D Space and Time (NeurIPS 2026 · similarity 0.84)
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.
