DeepBranchTracer: A Generally-Applicable Approach to Curvilinear Structure Reconstruction Using Multi-Feature Learning
This addresses the challenge of reconstructing line-like structures in computer vision applications, but it appears incremental as it builds on existing segmentation and reconstruction approaches with a multi-feature learning twist.
The paper tackled the problem of reconstructing curvilinear structures from images, which is challenging due to complex topology and ambiguous evidence, by introducing DeepBranchTracer, a method that learns external image and internal geometric features; it demonstrated superior performance in accuracy and continuity over existing methods on 2D and 3D datasets.
Curvilinear structures, which include line-like continuous objects, are fundamental geometrical elements in image-based applications. Reconstructing these structures from images constitutes a pivotal research area in computer vision. However, the complex topology and ambiguous image evidence render this process a challenging task. In this paper, we introduce DeepBranchTracer, a novel method that learns both external image features and internal geometric characteristics to reconstruct curvilinear structures. Firstly, we formulate the curvilinear structures extraction as a geometric attribute estimation problem. Then, a curvilinear structure feature learning network is designed to extract essential branch attributes, including the image features of centerline and boundary, and the geometric features of direction and radius. Finally, utilizing a multi-feature fusion tracing strategy, our model iteratively traces the entire branch by integrating the extracted image and geometric features. We extensively evaluated our model on both 2D and 3D datasets, demonstrating its superior performance over existing segmentation and reconstruction methods in terms of accuracy and continuity.