Abstract: In an embodiment an apparatus includes a processor configured to generate a feature extraction module using a dataset in which an attribute for each object is defined, receive an image obtained by a camera, extract an attribute of an object of interest from the image using the learned feature extraction module, identify an object re-identification candidate group based on the extracted attribute of the object of interest and re-identify the object of interest based on the identified object re-identification candidate group.

Abstract: Methods and systems for time series analysis include generating a text summary of a time series using a first large language model (LLM) agent. A prompt is generated using a multi-modal encoder with the time series and the text summary as inputs. An event prediction is generated using a second LLM agent with the text summary and the prompt as inputs. An action is performed responsive to the event prediction.

Abstract: Disclosed herein is a metamaterial sensing platform. The metamaterial sensing platform includes: a sensing portion configured to capture a particle; and a terahertz sensor portion configured to sense the particle by allowing a terahertz electromagnetic wave to be incident on the sensing portion; wherein the sensing portion includes a base substrate and a particle capture portion disposed on the base substrate and having a plurality of slits formed therein through which the particle is captured, and wherein a nanodisk reactive with the particle is disposed in the plurality of slits.

Abstract: A device, and method thereof, for training a model for human identification are provided may include a primary training module configured to primarily train the model with respect to a pre-prepared source dataset, a target subset generation module configured to generate a target subset by selecting some cameras from among a plurality of cameras included on a service robot, a feature vector extraction module configured to extract feature vectors of the target subset by using the model, a labeling module configured to perform labeling on the feature vectors, and a secondary training module configured to secondarily train the model with respect to a target dataset by using results of the labeling.

Abstract: Systems and methods for time-series forecasting via multi-modal augmentation and fusion. Time-series data and modality data can be decomposed into seasonal and trend representations with trend-seasonal decomposition. Using an encoder transformer model, time-series data embeddings and modality data embeddings can be concatenated from the seasonal representations and the trend representations to obtain crossed representations. Using the encoder transformer model, the modality data embeddings and the time-series data embeddings can be processed separately to obtain singular representations. The crossed representations and the singular representations can be augmented through joint trend-seasonal decomposition to obtain augmented seasonal data and augmented trend data. Using a decoder, augmented seasonal data and augmented trend data can be fused to obtain fused augmented data.

Abstract: The object of the present invention is to provide a nanogap-based hydrogen sensor in which a nanogap is formed on a substrate and a trace amount of hydrogen gas leakage can be detected in a non-contact manner using the nanogap, and a method of manufacturing the same. In order to achieve the above object, the present invention is characterized by including: a metamaterial manufactured by forming a metal nanoslot pattern on a wafer; a catalyst layer deposited on the surface of the metamaterial to form a nanogap inside the metal nanoslot; and a protective layer formed on the catalyst layer to protect the catalyst layer.

Abstract: Provided is a security code including a substrate, metamaterials on the substrate, a signal modulation pattern on the metamaterials, and a capping layer covering the signal modulation pattern and the metamaterials, wherein the metamaterials include a pair of metal patterns facing each other, the signal modulation pattern covers a portion of the metal patterns, and expose remaining of the metal patterns, and the signal modulation pattern has a different material from each of the metal patterns.

Abstract: In an embodiment an apparatus includes a processor configured to generate a feature extraction module using a dataset in which an attribute for each object is defined, receive an image obtained by a camera, extract an attribute of an object of interest from the image using the learned feature extraction module, identify an object re-identification candidate group based on the extracted attribute of the object of interest and re-identify the object of interest based on the identified object re-identification candidate group.

Abstract: An apparatus and a method for re-identifying an object to improve object re-identification performance using attribute information of the object are provided. The apparatus trains an object representation extraction model to train attribute representations, inputs an image obtained by a camera to the trained object representation extraction model, extracts object representations from the image using the trained object representation extraction model, and performs object re-identification based on the object representations.

Abstract: A roll joint is provided. The roll joint may comprise: a first body; a second body disposed opposite to and spaced apart from the first body; a connection body which is disposed between the first body and the second body and is not in contact with the first body and the second body; and a connection wire member which connects the first body, the second body, and the connection body to one another so as to make a tensegrity structure.

Abstract: An artificial joint is provided. The artificial joint may comprise: a first joint member including a first bone replacement part, and a (1-1)st branch and a (1-2)st branch branched from opposite sides of the first bone replacement part; a second joint member including a second bone replacement part, and a (2-1)nd branch and a (2-2)nd branch branched from opposite sides of the second bone replacement part; a first main string connecting one side of the (1-1)st branch and one side of the (2-1)nd branch; a second main string connecting the one side of the (1-1)st branch and one side of the (2-2)nd branch; a third main string connecting one side of the (1-2)st branch and the one side of the (2-1)nd branch; and a fourth main string connecting the one side of the (1-2)st branch and the one side of the (2-2)nd branch.

Abstract: A roll joint is provided. The roll joint may comprise: a first body; a second body disposed opposite to and spaced apart from the first body; a connection body which is disposed between the first body and the second body and is not in contact with the first body and the second body; and a connection wire member which connects the first body, the second body, and the connection body to one another so as to make a tensegrity structure.

Abstract: An artificial joint is provided. The artificial joint may comprise: a first joint member including a first bone replacement part, and a (1-1)st branch and a (1-2)st branch branched from opposite sides of the first bone replacement part; a second joint member including a second bone replacement part, and a (2-1)nd branch and a (2-2)nd branch branched from opposite sides of the second bone replacement part; a first main string connecting one side of the (1-1)st branch and one side of the (2-1)nd branch; a second main string connecting the one side of the (1-1)st branch and one side of the (2-2)nd branch; a third main string connecting one side of the (1-2)st branch and the one side of the (2-1)nd branch; and a fourth main string connecting the one side of the (1-2)st branch and the one side of the (2-2)nd branch.