WaspMOT: A Benchmark for Long-Term Multi-Object Tracking of Trichogramma Wasps

2026-07-09

AI Summary

This work addresses the limitation of existing multi-object tracking (MOT) benchmarks, which predominantly rely on short video clips and thus inadequately evaluate long-term identity consistency. To bridge this gap, we introduce the first closed-set, long-duration MOT benchmark tailored for the parasitic wasp *Trichogramma*, comprising 10 densely annotated videos of approximately 8 minutes each, with a focus on maintaining stable identity associations over thousands of frames. Under a unified evaluation protocol, we assess prominent track-by-detection methods such as ByteTrack and BoT-SORT, employing oracle detections to isolate association performance from detection errors. Our experiments reveal significant trajectory fragmentation across all baselines, yet demonstrate that even a simple spatial trajectory stitching strategy substantially improves long-term tracking accuracy, highlighting a critical deficiency—and considerable room for improvement—in current approaches regarding sustained identity preservation.

Abstract

Multi-object tracking (MOT) has achieved strong performance on benchmarks dominated by short video sequences. However, such datasets do not adequately evaluate long-term identity preservation, where objects must be tracked consistently over extended durations. We introduce WaspMOT, a benchmark designed to address this gap through long-duration tracking of Trichogramma wasps in controlled ecological experiments. The dataset contains 10 sequences of approximately 12,000 frames each (over 8 minutes at 25 FPS), with dense MOTChallenge annotations and oracle detections to isolate association performance. Unlike existing benchmarks, WaspMOT forms a closed-set tracking scenario where all individuals remain present throughout the sequence, requiring consistent identity assignment across thousands of frames despite abrupt jumps, occlusions, and highly similar appearance. We establish a benchmark by evaluating five tracking-by-detection methods, including ByteTrack, BoT-SORT, C-BIoU, OC-SORT, and McByte, under a unified protocol. Results show that all methods suffer from significant trajectory fragmentation, highlighting the difficulty of long-term identity preservation even with perfect detections. A simple spatial tracklet stitching baseline consistently improves performance, indicating that substantial gains remain possible. WaspMOT provides a new benchmark for studying long-term association and reveals limitations of current tracking approaches that are not observable on conventional datasets. The benchmark will be made publicly available at the project repository: https://github.com/tstanczyk95/WaspMOT/ .

Problem

• long-term multi-object tracking
• identity preservation
• tracking benchmark
• Trichogramma wasps
• trajectory fragmentation

Innovation

• long-term multi-object tracking
• identity preservation
• closed-set tracking
• trajectory fragmentation
• benchmark dataset