Batch-MOT: Batch-Enabled Real-Time Scheduling for Multi-Object Tracking Tasks

초록

Targeting a MOT (Multi-Object Tracking) system with multiple MOT tasks, this paper develops Batch-MOT, the first system design that achieves both (G1) timing guarantee and (G2) accuracy maximization, by utilizing batch execution that allows multiple DNN executions to perform simultaneously in a single DNN inference, resulting in significantly decreased execution time without accuracy loss. To this end, we propose an adaptable scheduling framework that allows run-time execution behaviors deviated from our base scheduling algorithm (i.e., nonpreemptive fixed-priority scheduling) without compromising G1. Based on the adaptable framework, we then develop (i) a runtime batching mechanism that finds and executes a batch set of MOT tasks and (ii) a run-time idling mechanism that waits for future releases of MOT tasks for batch execution. Both runtime mechanisms can achieve G1 and G2 without incurring high run-time overhead, as they systematically exploit the run-time execution behaviors allowed by the adaptive framework. Our evaluation conducted with a real-world data set demonstrates the effectiveness of Batch-MOT in improving tracking accuracy while providing a timing guarantee, compared to the state-of-the-art real-time MOT system for multiple MOT tasks. IEEE

키워드