Generating high-quality novel views of a scene from a single image requires maintaining structural coherence across different views, referred to as view consistency. While diffusion models have driven advancements in novel view synthesis, they still struggle to preserve spatial continuity across views. Diffusion models have been combined with 3D models to address the issue, but such approaches lack efficiency due to their complex multi-step pipelines. This paper proposes a novel view-consistent image generation method which utilizes diffusion models without additional modules. Our key idea is to enhance diffusion models with a training-free method that enables adaptive attention manipulation and noise reinitialization by leveraging view-guided warping to ensure view consistency. Through our comprehensive metric framework suitable for novel-view datasets, we show that our method improves view consistency across various diffusion models, demonstrating its broader applicability.

Disentangled representation learning (DRL) aims to break down observed data into core intrinsic factors for a profound understanding of the data. In real-world scenarios, manually defining and labeling these factors are non-trivial, making unsupervised methods attractive. Recently, there have been limited explorations of utilizing diffusion models (DMs), which are already mainstream in generative modeling, for unsupervised DRL. They implement their own inductive bias to ensure that each latent unit input to the DM expresses only one distinct factor. In this context, we design Dynamic Gaussian Anchoring to enforce attribute-separated latent units for more interpretable DRL. This unconventional inductive bias explicitly delineates the decision boundaries between attributes while also promoting the independence among latent units. Additionally, we also propose Skip Dropout technique, which easily modifies the denoising U-Net to be more DRL-friendly, addressing its uncooperative nature with the disentangling feature extractor. Our methods, which carefully consider the latent unit semantics and the distinct DM structure, enhance the practicality of DM-based disentangled representations, demonstrating state-of-the-art disentanglement performance on both synthetic and real data, as well as advantages in downstream tasks.

The risks arising from precipitation are immediate and broad. The Korea Meteorological Administration also carries out predictions and observations on precipitation in Korea. Still, the results are inaccurate due to cases where precipitation in the north is not well matched or incorrect observations are made due to difficulties in identifying the surface. We intend to construct accurate precipitation data by analyzing a machine-learning model. If highly influential precipitation data are properly secured, it is expected that it will benefit greatly from more accurate analysis of health effects.After collecting and refining data from 2015 to 2023 for about 50 detailed variables related to temperature, river, and snow using Google Earth Engine data, which is the input of the model, ASOS (Automated Synchronous Observing System) data from the Korea Meteorological Administration are collected in the same manner as in the relevant year and combined based on GeoID. Among many ASOS data parameters, precipitation data were used and provided as ground truth for the model. Random Forest, XGB (Extreme Gradient Boosting), which is known to be effective with those data, is used and the best performance model is chosen as the final model.The R2 of the model using XGB was about 0.8.When the additional hyperparameter was not used, the R2 of the model using XGB was the best at 0.8 which is a high value compared to the others

Having a seamless connection between independently developed platforms has recently become an important issue. In addition, research is underway on auto-configuration to minimize human interference while configuring devices for users to utilize Internet of Things (IoT) applications. OneM2M is a standardization group established to develop a standard Internet service platform. In the oneM2M environment, data exchange between each component is performed through HTTP, constrained application protocol (CoAP), and message queuing telemetry transport (MQTT). MQTT enables efficient data transmission in low-power and unreliable networks. However, MQTT is not suitable for service discovery for auto-configuration given that it does not support multicasting and resource directories for managing IoT resources. In this paper, we present an architecture to provide an auto-configuration mechanism by using the MQTT publish/subscribe messaging pattern. To realize fully autonomous systems, the proposed system autonomously interacts with devices by recognizing the context using semantic Web technology. We verify the feasibility of the proposed system through its implementation in the real environment. It is confirmed that the proposed approach is able to automatically configure MQTT-based devices and provide context-aware services using a standard interface.

Cohesion significantly impacts on posterior development and maintenance. A well-designed module should be coherent with a single responsibility and less propagative for changes to other modules. In object-oriented development, classes are often modularized based on their relationships to localize change impact. There has been much work on grouping classes using direct class interactions and similarity of lexical information of classes. However, there has been little attention to indirect class relationships which are significant in defining functional semantics in object-oriented systems. In this paper, we present an approach for grouping classes using both direct and indirect class relationships. In the approach, we define a set of metrics for measuring couplings based on indirect class relationships and use them to group classes with the aim of localizing the impact of changes. We evaluate the approach using three applications - JHotDraw, JMeter, and ArgoUML and compare the results with existing work. The evaluation shows that the proposed approach reduces 17.2% and 22.2% of the number of packages that are affected by changes over the existing works.