19 July 2024 in ACM Transactions on Graphics
Visual In-Context Learning (ICL) has emerged as a promising research area due to its capability to accomplish various tasks with limited example pairs through analogical reasoning. However, training-based visual ICL has limitations in its ability to generalize to unseen tasks and requires the collection of a diverse task dataset. On the other hand, existing methods in the inference-based visual ICL category solely rely on textual prompts, which fail to capture fine-grained contextual information from given examples and can be time-consuming when converting from images to text prompts. To address these challenges, we propose Analogist, a novel inference-based visual ICL approach that exploits both visual and textual prompting techniques using a text-to-image diffusion model pretrained for image inpainting. For visual prompting, we propose a self-attention cloning (SAC) method to guide the fine-grained structural-level analogy between image examples. For textual prompting, we leverage GPT-4V's visual reasoning capability to efficiently generate text prompts and introduce a cross-attention masking (CAM) operation to enhance the accuracy of semantic-level analogy guided by text prompts. Our method is out-of-the-box and does not require fine-tuning or optimization. It is also generic and flexible, enabling a wide range of visual tasks to be performed in an in-context manner. Extensive experiments demonstrate the superiority of our method over existing approaches, both qualitatively and quantitatively. Our project webpage is available at https://analogist2d.github.io.
27 June 2011 in Spectroscopy Letters
Calcium phosphate glasses, in which part of calcium oxide was replaced by lanthanum oxide, were prepared by using the conventional melt quench method. The structures of xLa2O3 · (50-x)CaO · 50P2O5 (x = 0, 1, 3, 6, 12 mol%) samples were investigated by X-ray diffraction (XRD), Raman spectrum, Fourier transform infrared spectrum (FTIR), and differential scanning calorimetry (DSC). The results show that lanthanum oxide acts as network modifier in the network space of glass structure. The glass formation occurs at an O/P ratio of about 3.0–3.12. At larger values, the crystalline phases calcium pyrophosphate (Ca2P2O7) and calcium lanthanum phosphate [Ca9La(PO4)7] are detected in the samples. Raman and FTIR spectra indicate that the structure of lanthanum-free sample is chain P–O–P bond metaphosphate–based Q2 units. Glass structure will change to Q2 and Q1 units when the lanthanum oxide content is less then 6 mol%. When lanthanum oxide content increases to 9 and 12 mol% more nonbridging oxygen in the glass, resulting in the depolymerization of the phosphate network, the network of glass transforms to Q2, Q1, and Q0 units mixture. Based upon DSC results, Tg slightly decreases because of the depolymerization of microstructure. Endothermal peak of DSC curves indicate that crystal phases separate out from vitreous body with the addition of lanthanum oxide content.
1 June 1992 in Japanese Journal of Applied Physics
Experimental evidence shows that the noise current is proportional to the flowing current. The samples are the P-N junction, and light-emitting and laser diodes. Noise analysis is carried out using Hooge's equation. The low-frequency noise agrees qualitatively with the analytical result.
10 December 1996 in Applied Optics
Optics InfoBase is the Optical Society's online library for flagship journals, partnered and copublished journals, and recent proceedings from OSA conferences.