Synthetic data enables faster annotation and robust segmentation for multi-object grasping in clutter
This work addresses the challenge of data scarcity and annotation costs for robotic grasping in cluttered environments, offering an incremental improvement over existing methods.
The paper tackles the problem of time-consuming and costly data annotation for robotic grasping by proposing a synthetic data generation method that reduces labeling time and improves segmentation robustness. The results show that using a hybrid dataset (synthetic plus real) improves labeling success rate by 6.7% and grasping success rate by 18.8% compared to a real dataset alone.
Object recognition and object pose estimation in robotic grasping continue to be significant challenges, since building a labelled dataset can be time consuming and financially costly in terms of data collection and annotation. In this work, we propose a synthetic data generation method that minimizes human intervention and makes downstream image segmentation algorithms more robust by combining a generated synthetic dataset with a smaller real-world dataset (hybrid dataset). Annotation experiments show that the proposed synthetic scene generation can diminish labelling time dramatically. RGB image segmentation is trained with hybrid dataset and combined with depth information to produce pixel-to-point correspondence of individual segmented objects. The object to grasp is then determined by the confidence score of the segmentation algorithm. Pick-and-place experiments demonstrate that segmentation trained on our hybrid dataset (98.9%, 70%) outperforms the real dataset and a publicly available dataset by (6.7%, 18.8%) and (2.8%, 10%) in terms of labelling and grasping success rate, respectively. Supplementary material is available at https://sites.google.com/view/synthetic-dataset-generation.