The scarcity of multimodal datasets in remote sensing, particularly those combining high-resolution imagery with descriptive textual annotations, limits advancements in context-aware analysis. To address this, we introduce a novel dataset comprising 12,473 aerial and satellite images sourced from established benchmarks (RSSCN7, DLRSD, iSAID, LoveDA, and WHU), enriched with automatically generated pseudo-captions and semantic tags.
The GestDoor dataset contains wearable sensor data collected to support research in biometric authentication through arm movements during door-opening interactions. Using two 6-degree-of-freedom (6-DOF) inertial measurement units (IMUs) worn on the wrist and upper arm, 11 participants performed four types of door-opening tasks—left-hand pull, left-hand push, right-hand pull, and right-hand push—across up to three sessions. The dataset includes 3,330 samples comprising accelerometer and gyroscope signals at 100 Hz, along with session metadata.
The data include multispectral data from landsat8 which was processed in ArcMap. METRIC model was used to calculate evapotranspiration. The data is then transferred to R for further processing and ANN model training.
This opinion explores the integration of Artificial Intelligence in foreign language education, examining both its potential benefits and inherent risks. AI tools offer personalized learning experiences, interactive practice, and access to authentic resources, potentially reducing learning-related stress. However, over-reliance on AI may hinder critical thinking and raise concerns about accuracy, originality, and ethical considerations like algorithmic bias. A balanced approach is crucial, emphasizing the importance of academic integrity, ethical conduct, and responsible technology use.
Software Information System (static & dynamic) analysis of entities
Wi-Fi sensing has emerged as a promising non-intrusive approach for human activity recognition (HAR), leveraging the widespread availability of Wi-Fi-enabled devices. While machine learning tools help extract meaningful features from complex wireless propagation patterns, developing robust and domain-adaptive models remains challenging without diverse datasets.
In the modern era of fast-paced lifestyles and increasing psychological pressure, real-time stress identification has become a critical component of mental health monitoring systems.
The video demonstration corresponding to the 100th time step in Figure 13 for the HalfCheetah controlled by the random policy and the learned
policies with different methods. MDDPG(5) denotes the model-free counterpart with 5-step TD target. FNN-Model-MDDPG(5) and ResNet-Model-MDDPG(5) denote the FNN-model-based and our ResNet-model-based schemes with 5 dynamics models, respectively.
Automatic Identification System (AIS) data are critical for maritime domain awareness, enabling tasks such as vessel classification, trajectory prediction, and anomaly detection. However, AIS datasets frequently suffer from domain shifts, data sparsity, and class imbalance, which limit the generalization of predictive models. To address these challenges, this paper presents AISCycleGen, a novel data augmentation framework that leverages Cycle-Consistent Generative Adversarial Networks to synthesize realistic AIS sequences through unpaired domain translation.
We developed IIST BCI Dataset-9, a novel EEG-based Brain-Computer Interface (BCI)
dataset to improve wheelchair control systems using Malayalam dialect variations. BCI
systems help people with motor disabilities by allowing them to control devices using brain
signals. The limited number of BCI datasets in Indian languages makes it harder for native
speakers to use these systems. To address this, we created a dataset with 15 Malayalam
words related to basic wheelchair commands like Forward, Backward, Go, Stop, Reverse,
