Search papers, labs, and topics across Lattice.
D pattern contours. AutoSew formulates the problem as a graph matching task, leveraging a Graph Neural Network to capture local and global geometric context, and employing a differentiable optimal transport solver to infer stitching relationships鈥攊ncluding multi-edge connections. To support this task, we update the GarmentCodeData dataset modifying over 18k patterns with realistic multi-edge annotations, reflecting industrial assembly scenarios. AutoSew achieves 96% F1-score and successfully assembles 73.3% of test garments without error, outperforming existing methods while relying solely on geometric input. Our results demonstrate that geometry alone can robustly guide stitching prediction, enabling scalable garment assembly without manual input. Our dataset and code are available online. 111https://mslab.es/projects/autosew/
2
0
2
Key contribution not extracted.
Ditch manual annotations: AutoSew uses only 2D pattern geometry and GNNs to predict garment stitching with high accuracy, even with complex multi-edge connections.
