Adaptive Subsampling for ROI-based Visual Tracking: Algorithms and FPGA Implementation

TL;DR

This paper introduces adaptive subsampling algorithms for energy-efficient ROI-based visual tracking, combining object detection and prediction, implemented on FPGA, achieving competitive accuracy and power efficiency.

Contribution

It presents a novel adaptive subsampling approach for visual tracking that integrates ROI prediction with FPGA implementation, enhancing energy efficiency and real-time performance.

Findings

Coupling ECO tracker with Kalman filter yields AUC scores of 0.4568 (OTB100) and 0.3471 (LaSOT).

The FPGA implementation achieves near-real-time tracking at 19.23 FPS.

Power consumption is approximately 4 W for ECO-based and 6 W for YOLO-based algorithms.

Abstract

There is tremendous scope for improving the energy efficiency of embedded vision systems by incorporating programmable region-of-interest (ROI) readout in the image sensor design. In this work, we study how ROI programmability can be leveraged for tracking applications by anticipating where the ROI will be located in future frames and switching pixels off outside of this region. We refer to this process of ROI prediction and corresponding sensor configuration as adaptive subsampling. Our adaptive subsampling algorithms comprise an object detector and an ROI predictor (Kalman filter) which operate in conjunction to optimize the energy efficiency of the vision pipeline with the end task being object tracking. To further facilitate the implementation of our adaptive algorithms in real life, we select a candidate algorithm and map it onto an FPGA. Leveraging Xilinx Vitis AI tools, we designed and accelerated a YOLO object detector-based adaptive subsampling algorithm. In order to further improve the algorithm post-deployment, we evaluated several competing baselines on the OTB100 and LaSOT datasets. We found that coupling the ECO tracker with the Kalman filter has a competitive AUC score of 0.4568 and 0.3471 on the OTB100 and LaSOT datasets respectively. Further, the power efficiency of this algorithm is on par with, and in a couple of instances superior to, the other baselines. The ECO-based algorithm incurs a power consumption of approximately 4 W averaged across both datasets while the YOLO-based approach requires power consumption of approximately 6 W (as per our power consumption model). In terms of accuracy-latency tradeoff, the ECO-based algorithm provides near-real-time performance (19.23 FPS) while managing to attain competitive tracking precision.