Research map / Detection and segmentation / Tracking
Visual Object Tracking in First Person Vision
Matteo Dunnhofer, Antonino Furnari, Giovanni Maria Farinella, Christian Micheloni
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 LTMU-H (hand-based re-detection) scheme | DiMP-MU · TREK-150 OPE | SS: 0.411 -> 0.461 | Improved |
| add LTMU-H scheme | DiMP-MU · TREK-150 OPE | GSR: 0.320 -> 0.376 | Improved |
| add LTMU-H scheme | STMTrack · TREK-150 OPE | SS: 0.464 -> 0.487 | Improved |
| add LTMU-H scheme | STMTrack · TREK-150 OPE | GSR: 0.451 -> 0.438 | Hurt |
| add LTMU-F scheme | STARK · TREK-150 OPE | GSR: 0.395 -> 0.334 | Hurt |
| add LTMU-H scheme | STARK · TREK-150 OPE | SS: 0.492 -> 0.505 | Improved |
| add LTMU-F scheme | STMTrack · TREK-150 OPE | SS: 0.464 -> 0.461 | No clear effect |
| replace SORT with IoU-with-previous-box association (ablation) | TbyD-H · TREK-150 OPE | SS: 0.237 -> 0.433 | Improved |
| replace SORT with IoU-with-previous-box association (ablation) | TbyD-F · TREK-150 OPE | SS: 0.313 -> 0.390 | Improved |
| add SORT on top of IoU-with-previous-box association (ablation) | TbyD-H · TREK-150 OPE | SS: 0.433 -> 0.432 | No clear effect |
| initialize trackers with Faster-R-CNN detection instead of ground truth | generic trackers · TREK-150 OPE-D | SS, NPS, GSR change: -5.3%, -5.1%, -3.1% | Hurt |
| initialize trackers with HiC detection instead of ground truth | generic trackers · TREK-150 OPE-D | SS, NPS, GSR change: -5.9%, -5.7%, -4.4% | Hurt |
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.
- STMTrack — STMTrack: Template-Free Visual Tracking With Space-Time Memory Networks (CVPR 2021) · better on 2 of 3 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
- Is Tracking Really More Challenging in First Person Egocentric Vision? (ICCV 2025 · similarity 0.89)
- Beyond Local Search: Tracking Objects Everywhere With Instance-Specific Proposals (CVPR 2016 · similarity 0.89)
- OmniTracker: Unifying Visual Object Tracking by Tracking-With-Detection (TPAMI 2025 · similarity 0.88)
- Beyond Accuracy: Tracking more like Human via Visual Search (NeurIPS 2024 · similarity 0.88)
- OVT-B: A New Large-Scale Benchmark for Open-Vocabulary Multi-Object Tracking (NeurIPS 2024 · similarity 0.87)
- TrackIME: Enhanced Video Point Tracking via Instance Motion Estimation (NeurIPS 2024 · similarity 0.87)
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.
