Carton dataset synthesis method for domain shift based on foreground texture decoupling and replacement
It addresses domain shift in industrial object detection, enabling better model generalization across different logistics and market scenarios, but is incremental as it builds on existing synthesis techniques.
The paper tackles the problem of domain shift in object detection for industrial carton datasets by proposing a novel image synthesis method that replaces foreground textures from source to target domains while preserving context relationships. This method boosts AP by 4.3% to 6.8% on models like RetinaNet and Faster R-CNN for target domains.
One major impediment in rapidly deploying object detection models for industrial applications is the lack of large annotated datasets. We currently have presented the Sacked Carton Dataset(SCD) that contains carton images from three scenarios, such as comprehensive pharmaceutical logistics company(CPLC), e-commerce logistics company(ECLC), fruit market(FM). However, due to domain shift, the model trained with one of the three scenarios in SCD has poor generalization ability when applied to the rest scenarios. To solve this problem, a novel image synthesis method is proposed to replace the foreground texture of the source datasets with the texture of the target datasets. Our method can keep the context relationship of foreground objects and backgrounds unchanged and greatly augment the target datasets. We firstly propose a surface segmentation algorithm to achieve texture decoupling of each instance. Secondly, a contour reconstruction algorithm is proposed to keep the occlusion and truncation relationship of the instance unchanged. Finally, the Gaussian fusion algorithm is used to replace the foreground texture from the source datasets with the texture from the target datasets. The novel image synthesis method can largely boost AP by at least 4.3%~6.5% on RetinaNet and 3.4%~6.8% on Faster R-CNN for the target domain. Code is available at https://github.com/hustgetlijun/RCAN.