Search

Using density functional theory calculations, atomic and electronic structure of defects in monolayer GeS were investigated by focusing on the effects of vacancies and substitutional atoms. We chose group IV or chalcogen elements as substitutional ones, which substitute for Ge or S in GeS. It was found that the bandgap of GeS with substitutional atoms is close to that of pristine GeS, while the bandgap of GeS with Ge or S vacancies was smaller than that of pristine GeS. In terms of formation energy, monolayer GeS with Ge vacancies is more stable than that with S vacancies, and notably GeS with Ge substituted with Sn is most favorable within the range of chemical potential considered. Defects affect the piezoelectric properties depending on vacancies or substitutional atoms. Especially, GeS with substitutional atoms has almost the same piezoelectric stress coefficients eij as pristine GeS while having lower piezoelectric strain coefficients dij but still much higher than other 2D materials. It is therefore concluded that Sn can effectively heal Ge vacancy in GeS, keeping high piezoelectric strain coefficients.

Recent Transformer-based object detectors have achieved remarkable performance on benchmark datasets, but few have addressed the real-world challenge of object detection in crowded scenes using transformers. This limitation stems from the fixed query set size of the transformer decoder, which restricts the model’s inference capacity. To overcome this challenge, we propose Recurrent Detection Transformer (Recurrent DETR), an object detector that iterates the decoder block to render more predictions with a finite number of query tokens. Recurrent DETR can adaptively control the number of decoder block iterations based on the image’s crowdedness or complexity, resulting in a variable-size prediction set. This is enabled by our novel Pondering Hungarian Loss, which helps the model to learn when additional computation is required to identify all the objects in a crowded scene. We demonstrate the effectiveness of Recurrent DETR on two datasets: COCO 2017, which represents a standard setting, and CrowdHuman, which features a crowded setting. Our experiments on both datasets show that Recurrent DETR achieves significant performance gains of 0.8 AP and 0.4 AP, respectively, over its base architectures. Moreover, we conduct comprehensive analyses under different query set size constraints to provide a thorough evaluation of our proposed method.