DCFNet: Discriminant Correlation Filters Network for Visual Tracking

Abstract

Discriminant Correlation Filters (DCF) based methods now become a kind of dominant approach to online object tracking. The features used in these methods, however, are either based on hand-crafted features like HoGs, or convolutional features trained independently from other tasks like image classification. In this work, we present an end-to-end lightweight network architecture, namely DCFNet, to learn the convolutional features and perform the correlation tracking process simultaneously. Specifically, we treat DCF as a special correlation filter layer added in a Siamese network, and carefully derive the backpropagation through it by defining the network output as the probability heatmap of object location. Since the derivation is still carried out in Fourier frequency domain, the efficiency property of DCF is preserved. This enables our tracker to run at more than 60 FPS during test time, while achieving a significant accuracy gain compared with KCF using HoGs. Extensive evaluations on OTB-2013, OTB-2015, and VOT2015 benchmarks demonstrate that the proposed DCFNet tracker is competitive with several state-of-the-art trackers, while being more compact and much faster.

5 pages, 4 figures

References in corpus (3)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Understanding and Diagnosing Visual Tracking Systems
• On the Computation of Complex-valued Gradients with Application to Statistically Optimum Beamforming

Cited by in corpus (30)

• GOT-10k: A Large High-Diversity Benchmark for Generic Object Tracking in the Wild
• Deep Learning for Visual Tracking: A Comprehensive Survey
• Good Features to Correlate for Visual Tracking
• SiamRPN++: Evolution of Siamese Visual Tracking with Very Deep Networks
• Visual Object Tracking in First Person Vision
• Do Different Tracking Tasks Require Different Appearance Models?
• Unsupervised Deep Tracking
• TRAT: Tracking by Attention Using Spatio-Temporal Features
• End-to-end Flow Correlation Tracking with Spatial-temporal Attention
• Multi-hierarchical Independent Correlation Filters for Visual Tracking
• Deep Convolutional Correlation Iterative Particle Filter for Visual Tracking
• A Review of Visual Trackers and Analysis of its Application to Mobile Robot
• Single-Camera Basketball Tracker through Pose and Semantic Feature Fusion
• Unsupervised Deep Representation Learning for Real-Time Tracking
• Improving Model Drift for Robust Object Tracking
• DensSiam: End-to-End Densely-Siamese Network with Self-Attention Model for Object Tracking
• P2B: Point-to-Box Network for 3D Object Tracking in Point Clouds
• An Efficient UAV-based Artificial Intelligence Framework for Real-Time Visual Tasks
• Towards a Better Match in Siamese Network Based Visual Object Tracker
• High Performance Visual Object Tracking with Unified Convolutional Networks
• An Exploration of Target-Conditioned Segmentation Methods for Visual Object Trackers
• Beyond Background-Aware Correlation Filters: Adaptive Context Modeling by Hand-Crafted and Deep RGB Features for Visual Tracking
• DSNet: Deep and Shallow Feature Learning for Efficient Visual Tracking
• Is First Person Vision Challenging for Object Tracking?
• Real Time Visual Tracking using Spatial-Aware Temporal Aggregation Network
• Model-free Tracking with Deep Appearance and Motion Features Integration
• Handcrafted and Deep Trackers: Recent Visual Object Tracking Approaches and Trends
• MobiFace: A Novel Dataset for Mobile Face Tracking in the Wild
• Generic Multiview Visual Tracking
• Exemplar Loss for Siamese Network in Visual Tracking