Deep-LK for Efficient Adaptive Object Tracking

In this paper, we present a new approach for efficient regression-based object tracking. Our approach is closely related to the Generic Object Tracking Using Regression Networks (GOTURN) framework [1]. We make the following contributions. First, we demonstrate that there is a theoretical relationship between Siamese regression networks like GOTURN and the classical Inverse Compositional Lucas & Kanade (IC-LK) algorithm. Further, we demonstrate that unlike GOTURN, IC-LK adapts its regressor to the appearance of the current tracked frame. We argue that the lack of such property in GOTURN attributes to its poor performance on unseen objects and/or viewpoints. Second, we propose a novel framework for object tracking inspired by the IC-LK framework, which we refer to as Deep-LK. Finally, we show impressive results demonstrating that Deep-LK substantially outperforms GOTURN and demonstrate comparable tracking performance against current state-of-the-art deep trackers on high frame-rate sequences whilst being an order of magnitude (100 FPS) computationally efficient.

Authors

• Chaoyang Wang 0001
• Hamed Kiani Galoogahi
• Chen-Hsuan Lin 0001
• Simon Lucey

Keywords

• Object tracking
• Mathematical model
• Robustness
• Feature extraction
• Linear regression
• Videos
• Strain
• Tracking Efficiency
• Computational Efficiency
• Regression Network
• Siamese Network
• Frames Per Second
• Unseen Objects
• Deep Learning
• Parametrized
• Intersection Over Union
• Bounding Box
• Unmanned Aerial Vehicles
• Deep Features
• Source Images
• Changes In Appearance
• AlexNet
• Motor Changes
• Appearance Variations
• Geometric Transformation
• Sum Of Squared Differences
• Template Image
• Correlation Filter
• Geometric Deformation
• Laplace Distribution
• Tracking Dataset
• Cost Curve
• Tracking Speed
• Convolutional Features
• Scale Changes
• Multilayer Perceptron