Co-PLNet: A Collaborative Point-Line Network for Prompt-Guided Wireframe Parsing

Co-PLNet is introduced as a novel framework for wireframe parsing that integrates point and line detection tasks through a Point-Line Prompt Encoder (PLP-Encoder) and a Cross-Guidance Line Decoder (CGL-Decoder). This architecture enables spatial cue exchange to improve prediction consistency and robustness, achieving enhanced accuracy and real-time efficiency in structured geometry perception, as demonstrated in experiments on the Wireframe and YorkUrban datasets. This approach is significant for practitioners as it addresses the limitations of existing methods by providing a more integrated solution for geometric representation in applications like SLAM.