Search for: All records

Few-shot instance segmentation extends the few-shot learning paradigm to the instance segmentation task, which tries to segment instance objects from a query image with a few annotated examples of novel categories. Conventional approaches have attempted to address the task via prototype learning, known as point estimation. However, this mechanism depends on prototypes (e.g. mean of K-shot) for prediction, leading to performance instability. To overcome the disadvantage of the point estimation mechanism, we propose a novel approach, dubbed MaskDiff, which models the underlying conditional distribution of a binary mask, which is conditioned on an object region and K-shot information. Inspired by augmentation approaches that perturb data with Gaussian noise for populating low data density regions, we model the mask distribution with a diffusion probabilistic model. We also propose to utilize classifier-free guided mask sampling to integrate category information into the binary mask generation process. Without bells and whistles, our proposed method consistently outperforms state-of-the-art methods on both base and novel classes of the COCO dataset while simultaneously being more stable than existing methods. The source code is available at: https://github.com/minhquanlecs/MaskDiff. 

Conjugated microporous polymers (CMPs) are porous organic materials that display (semi)conducting behavior due to their highly π-conjugated structures, making them promising next-generation materials for applications requiring both electrical conductivity and porosity. Currently, most CMPs and related porous aromatic frameworks (PAFs) are prepared using expensive transition metals (e.g., Pd), significantly increasing the costs associated with their synthesis. Lewis acid-mediated cyclotrimerization reactions of methyl ketones and nitriles represent promising and green alternative methods for CMP and PAF synthesis. Herein, we demonstrate that the generality of the solvent-free cyclotrimerization reactions is significantly improved by using ZnBr2 instead of ZnCl2 as the ionothermal medium. Specifically, we show that 1,4-diacetylbenzene (DAB), 4,4′-diacetylbiphenyl (DABP), 2,7-diacetylfluorene (DAF), 1,3,5-triacetylbenzene (TAB), tetrakis(4-acetylphenyl)methane (TAPM), and 1,4-dicyanobenzene (DCNB) can be polymerized in molten ZnBr2 to produce highly conjugated and microporous materials, as confirmed by 77 K N2 adsorption measurements, IR, and solid-state NMR. These findings support that ZnBr2 is an excellent Lewis acid mediator and medium for the ionothermal synthesis of porous organic materials.