Deep Feature Tracker: A Novel Application for Deep Convolutional Neural Networks

TL;DR

This paper introduces Deep-PT, a deep learning-based feature tracking method that learns to detect and track features reliably, outperforming existing algorithms especially in challenging environments like surgical images.

Contribution

The paper presents a unified deep neural network that simultaneously detects and tracks features, improving robustness and accuracy over prior methods.

Findings

Deep-PT outperforms recent pixel tracking algorithms in accuracy.

The network effectively detects reliable features for tracking.

Deep-PT demonstrates robustness against scene dynamics.

Abstract

Feature tracking is the building block of many applications such as visual odometry, augmented reality, and target tracking. Unfortunately, the state-of-the-art vision-based tracking algorithms fail in surgical images due to the challenges imposed by the nature of such environments. In this paper, we proposed a novel and unified deep learning-based approach that can learn how to track features reliably as well as learn how to detect such reliable features for tracking purposes. The proposed network dubbed as Deep-PT, consists of a tracker network which is a convolutional neural network simulating cross-correlation in terms of deep learning and two fully connected networks that operate on the output of intermediate layers of the tracker to detect features and predict trackability of the detected points. The ability to detect features based on the capabilities of the tracker distinguishes the proposed method from previous algorithms used in this area and improves the robustness of the algorithms against dynamics of the scene. The network is trained using multiple datasets due to the lack of specialized dataset for feature tracking datasets and extensive comparisons are conducted to compare the accuracy of Deep-PT against recent pixel tracking algorithms. As the experiments suggest, the proposed deep architecture deliberately learns what to track and how to track and outperforms the state-of-the-art methods.