Search for: All records

We propose a new approach for high resolution semantic image synthesis. It consists of one base image generator and multiple class-specific generators. The base generator generates high quality images based on a segmentation map. To further improve the quality of different objects, we create a bank of Generative Adversarial Networks (GANs) by separately training class-specific models. This has several benefits including – dedicated weights for each class; centrally aligned data for each model; additional training data from other sources, potential of higher resolution and quality; and easy manipulation of a specific object in the scene. Experiments show that our approach can generate high quality images in high resolution while having flexibility of object-level control by using class-specific generators. Project page: https://yuheng-li.github.io/CollageGAN/ 

Training generative models, such as GANs, on a target domain containing limited examples (e.g., 10) can easily result in overfitting. In this work, we seek to utilize a large source domain for pretraining and transfer the diversity information from source to target. We propose to preserve the relative similarities and differences between instances in the source via a novel cross-domain distance consistency loss. To further reduce overfitting, we present an anchor-based strategy to encourage different levels of realism over different regions in the latent space. With extensive results in both photorealistic and non-photorealistic domains, we demonstrate qualitatively and quantitatively that our few-shot model automatically discovers correspondences between source and target domains and generates more diverse and realistic images than previous methods. 

A fully integrated graphene field‐effect transistor (GFET) nanosensor utilizing a novel high‐κ solid‐gating geometry for a practical biosensor with enhanced sensitivity is presented. Herein, an “in plane” gate supplying electrical field through a 30 nm HfO₂dielectric layer is employed to eliminate the cumbrous external wire electrode in conventional liquid‐gate GFET nanosensors that undesirably limits the device potential in on‐site sensing applications. In addition to the advantage in the device integration degree, the transconductance level is found to be increased by about 50% over liquid‐gate GFET devices in aqueous‐media, thereby improves the sensitivity performance in sensor applications. As the first demonstration of biosensing applications, a small‐molecule antibiotic, kanamycin A, is detected by means of an aptameric competitive affinity principle. It is experimentally shown that the label‐free and specific quantification of kanamycin A with a concentration resolution at 11.5 × 10⁻⁹mis achievable through a single direct observation of the 200 s fast bioassay without any further noise canceling. These results demonstrate the utility and practicability of the new devices in label‐free biosensing as a novel analytical tool, and potentially hold great promise in other significant biomedical applications.