Abstract

The dataset contains the source code and model weights utilized for the experimental validation on segmentation of deformable linear objects with text prompts. The developed approach is called DLO Perceiver. The method employs the integration of language-based inputs to simplify the perception task of deformable linear objects. In particular, the input image is augmented with a text-based prompt guiding the segmentation of the target DLO. After encoding the image and text separately, a Perceiver-inspired structure is exploited to compress the concatenated data into transformer layers and generate the output mask from a latent vector representation. The data were produced in the framework of Horizon Europe IntelliMan project and were presented in the following publication: A. Caporali, K. Galassi and G. Palli, "DLO Perceiver: Grounding Large Language Model for Deformable Linear Objects Perception", in IEEE Robotics and Automation Letters, vol. 9, no. 12, pp. 11385-11392, Dec. 2024, doi: 10.1109/LRA.2024.3491428.