… Completion. We introduce a new single image GAN model targeted for the task of image
completion… Our model is trained on a single image with a missing region, and then suggests …

… Image completion aims at filling missing regions in images with plausibly synthesized
content. Example-based methods exploit the redundancy in natural images and transfer patches …

… Interestingly, FRAGSITE also identified a series of diverse scaffolds as true hit binders for
ACVR1 that could serve as a starting point for in silico and experimental fragment-based drug …

… In addition, we compare various fragment-based baseline models and the SOTA GNN … This
results in significantly improved performance compared to existing fragment-based and graph…

… In this work, since we aim to train an image completion model to generate objects, the
random masking process is not suitable. Therefore, we design a new data synthesis method that …

… image completion without taking into consideration the image masking as an inverse process
of completion… to an integrated framework of image completion and masking based on the …

… namely edge-based image inpainting to produce a complete image. For the discriminator
network, we use the concatenation of generated edgeness map and generated image as input, …

… model on small images with resolutions 512 x 512 and perform inference on high-resolution
images, achieving compelling inpainting quality. Our model can inpaint images as large as …

… introduce several fragmentbased contrastive and … fragment-based GNNs for downstream
prediction, thus utilizing the fragment information during finetuning. Our graph fragment-based …

… decades, fragment-based drug … fragment-based drug discovery, commonly used metrics
including ‘rule of three’ and ligand efficiency were developed [[11, 19]]. FBDD, fragment-based …