Abstract:Fast object tracking on embedded devices is of great importance for applications such as autonomous driving, unmanned aerial vehicle, and intelligent monitoring. Whereas, most of previous general solutions failed to reach this goal due to the facts that (i) high computational complexity and heterogeneous operation steps in the tracking models and (ii) parallelism-limited and bloated hardware platforms (e.g., CPU/GPU). Although previously proposed devices leverage neural dynamics and near-data processing for efficient tracking, their flexibility is limited due to the tight integration with vision sensor and the effectiveness on various video datasets is yet to be fully demonstrated. On the other side, recently the many-core architecture with massive parallelism and optimized memory locality is being widely applied to improve the performance for flexibly executing neural networks. This motivates us to adapt and map an object tracking model based on attractor neural networks with continuous and smooth attractor dynamics onto neural network chips for fast tracking. In order to make the model hardware friendly, we add local-connection restriction. We analyze the tracking accuracy and observe that the model achieves comparable results on typical video datasets. Then, we design a many-core neural network architecture with several computation and transformation operations to support the model. Moreover, by discretizing the continuous dynamics to the corresponding discrete counterpart, designing a slicing scheme for efficient topology mapping, and introducing a constant-restricted scaling chain rule for data quantization, we build a complete mapping framework to implement the tracking model on the many-core architecture. We fabricate a many-core neural network chip to evaluate the real execution performance. Results show that a single chip is able to accommodate the whole tracking model, and a fast tracking speed of nearly 800 FPS (frames per second) can be achieved. This work enables high-speed object tracking on embedded devices which normally have limited resources and energy.

Binarized Depthwise Separable Neural Network for Object Tracking in FPGA

Infrared Small Target Tracking Based on Sopc

A Fast and Energy Efficient FPGA-based System for Real-Time Object Tracking

Fast Object Tracking on a Many-Core Neural Network Chip

FPGA-Based Vehicle Detection and Tracking Accelerator

FPGA-based Acceleration System for Visual Tracking

Towards High-accuracy and Real-time Two-stage Small Object Detection on FPGA

Motion Object Tracking System Based on FPGA

P2M-DeTrack: Processing-in-Pixel-in-Memory for Energy-efficient and Real-Time Multi-Object Detection and Tracking

MiniTracker: A Lightweight CNN-based System for Visual Object Tracking on Embedded Device

Real-time low-power binocular stereo vision based on FPGA

Performance comparison of CNN, QNN and BNN deep neural networks for real-time object detection using ZYNQ FPGA node

Reduced-Parameter YOLO-Like Object Detector Oriented to Resource-Constrained Platform

Real-time implementation of fast discriminative scale space tracking algorithm

An FPGA Accelerator for High-Speed Moving Objects Detection and Tracking With a Spike Camera

Real-time tracking based on deep feature fusion

Design of a Real-Time Movement Decomposition-Based Rodent Tracker and Behavioral Analyzer Based on FPGA

Binary Neural Network in Robotic Manipulation: Flexible Object Manipulation for Humanoid Robot Using Partially Binarized Auto-Encoder on FPGA

Low Latency Edge Classification GNN for Particle Trajectory Tracking on FPGAs

A low-power end-to-end hybrid neuromorphic framework for surveillance applications

Accelerating Low Bit-Width Convolutional Neural Networks with Embedded FPGA.