The article introduces a new method of feature extraction for deformable linear objects. The method uses a Bezier curve to represent the original geometric shape, combined with a parameterization technique. The goal is to compute regression features from visual-feedback RGB images, capturing complex characteristics and providing an efficient shape feature in low-dimensional latent space. The method is evaluated through simulations, demonstrating better performance than existing methods in prediction accuracy, robustness, and computational efficiency. This work represents a significant advancement in the use of Bezier curve for shape representation.
Publication date: 29 Dec 2023
Project Page: Not Provided
Paper: https://arxiv.org/pdf/2312.16502
