Bibliographic record
Abstract
Multi-Object Tracking (MOT) is fundamental to applications such as autonomous driving, video surveillance, and sports analytics. However, challenges persist in maintaining long-term identity associations, handling dynamic object counts, managing irregular motion patterns, and mitigating occlusions in complex environments. Inspired by advancements in state-space models (SSMs), particularly Mamba [1], we propose a novel learning-based motion prediction architecture that integrates Mamba’s input-dependent sequence modeling with self-attention layers to effectively capture non-linear motion patterns within the Tracking-By-Detection (TBD) framework. Mamba’s ability to model long-range dependencies enhances motion prediction. We further refine spatial association by improving the cost matrix traditionally based on Intersection over Union (IoU). We incorporate Height-based IoU and extend bounding boxes using adjusted buffers to account for fast motion and partial overlaps, increasing robustness in object association. Achieving accurate MOT requires a balance between precise motion modeling and effective spatial and appearance matching, leveraging both strong and weak cues in data association. Our method is evaluated on challenging benchmarks, such as DanceTrack [2] and SportsMOT [3], achieving HOTA scores of 63.16% and 77.26%, respectively. These results surpass multiple state-of-the-art methods with a 3-7% improvement over other learning-based motion models, demonstrating the effectiveness of our approach in real-world tracking scenarios.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.003 | 0.002 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.004 | 0.002 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".