Trajectory-Aware Adaptive Inference in Object Detection Models
Quick Take
The study introduces a trajectory-aware adaptive inference method for efficient object detection in maritime navigation.
Key Points
- Incorporates GPS trajectory data for adaptive computation.
- Uses early-exit mechanism in YOLOv8-based detector.
- Achieves a trade-off between accuracy and efficiency.
📖 Reader Mode
~2 min readAbstract:The increasing integration of sensors in autonomous maritime navigation has led to large-scale multimodal datasets, raising challenges in achieving efficient real-time perception. In such systems, object detection and trajectory perception of nearby vessels are tightly coupled, particularly in dynamic environments such as maritime navigation. However, the efficiency of object detection models during inference remains an often-overlooked aspect. To this end, we build upon an existing object detection framework by incorporating GPS trajectory data into the inference process to enable input-adaptive computation. Specifically, we introduce an early-exit mechanism in a YOLOv8-based detector that incorporates motion cues - such as inter-vessel distances. Frames of vessels that are separated by short distances, converging with high speed, are processed using the full model, while only a subset of the network's architecture is activated otherwise. The difficulty degree (or scene complexity) of a frame or set of frames per second is evaluated by leveraging inter-object distance and the rate at which the distance between them decreases. Experimental results demonstrate that this strategy maintains satisfactory detection performance while significantly reducing inference time and computational cost, thus enabling a flexible trade-off between accuracy and efficiency compared to full-model inference.
| Comments: | Accepted to the MuseKDE workshop of the IEEE MDM 2026 conference |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2605.16397 [cs.CV] |
| (or arXiv:2605.16397v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2605.16397 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Grigorios Papanikolaou [view email]
[v1]
Tue, 12 May 2026 16:04:07 UTC (101 KB)
— Originally published at arxiv.org
More from arXiv cs.CV
See more →GeoSym127K: Scalable Symbolically-verifiable Synthesis for Multimodal Geometric Reasoning
GeoSym127K introduces a scalable neuro-symbolic framework for enhanced geometric reasoning in multimodal models.