Related Concept Videos
Aggregates Classification
Petrographic classification groups aggregates based on common mineralogical characteristics. Some of the common mineral groups found in aggregates are...
Random Variables
Uppercase letters such as X or Y denote a random variable. Lowercase letters like x or y denote the value of a random variable. If X is a random variable, then X is written in words, and x is given as a number.
For example, let X = the...
Prediction Intervals
However, the point estimate is most likely not the exact value of the population parameter, but close to it. After calculating point estimates, we construct interval estimates, called confidence intervals or prediction intervals. This prediction interval comprises a range of values unlike the point estimate and is a better predictor of the observed sample value, y.
Observational Learning
Force Classification
Contact and non-contact forces are two of the most widely used categories of forces. As the name suggests, contact forces require physical contact between two objects to act upon each other. Examples of contact forces include frictional,...
Per-Unit Sequence Models
Zero-sequence currents, which are identical in magnitude and phase, generate a neutral current, resulting in voltage drops across the neutral impedance and the low-voltage winding. If the...
You might also read
Related Articles
Articles linked to this work by shared authors, journal, and citation graph.
PDGrad: Guiding Diffusion Model for Reference-Based Blind Face Restoration with Pivot Direction Gradient Guidance.
ETFT: Equiangular Tight Frame Transformer for Imbalanced Semantic Segmentation.
ABDGAN: Arbitrary Time Blur Decomposition Using Critic-Guided TripleGAN.
Image Recommendation System Based on Environmental and Human Face Information.
Efficient Multi-Scale Stereo-Matching Network Using Adaptive Cost Volume Filtering.
Semantic Segmentation Using Pixel-Wise Adaptive Label Smoothing via Self-Knowledge Distillation for Limited Labeling Data.
RETRACTED: Zhang et al. A Novel Framework for Reconstruction and Imaging of Target Scattering Centers via Wide-Angle Incidence in Radar Networks. <i>Sensors</i> 2025, <i>25</i>, 6802.
Enhancing Unsupervised Multi-Source Domain Adaptation for Person Re-Identification via Mixture of Experts and Graph-Based Relation.
Development of an Instrumented Glove for Palmar Pressure Assessment in Kayakers.
Development and Experimental Validation of an Autonomous IoT-Based Monitoring System for Real-Time Water Quality Assessment in the Amazon River.
Semi-Supervised Adversarial Learning Framework for Controller Area Network Bus Intrusion Detection.
Smart Optimization Method for Safety Signs in Innovative Manufacturing Environments Integrating Industrial Field IoT Sensors and Knowledge Graphs.
Related Experiment Video
Updated: Jul 13, 2025

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
GammaGAN: Gamma-Scaled Class Embeddings for Conditional Video Generation.
Minjae Kang1, Yong Seok Heo1,2
1Department of Electrical and Computer Engineering, Ajou University, Suwon 16499, Republic of Korea.
GammaGAN improves conditional video generation by effectively using class labels via projection and normalization. This novel approach enhances video quality from single images, outperforming existing methods.
Area of Science:
- Computer Vision
- Artificial Intelligence
- Machine Learning
Background:
- Conditional video generation from single images with class labels is challenging.
- Traditional conditional generative adversarial networks (cGANs) struggle with effective class label utilization.
Purpose of the Study:
- To propose GammaGAN, a novel model for improved conditional video generation.
- To enhance the utilization of class labels in generative adversarial networks for video synthesis.
Main Methods:
- Developed GammaGAN with two streams: a class embedding stream and a data stream.
- Employed a projection method for effective class label utilization.
- Introduced scaling of class embeddings and normalization of data stream outputs to balance feature vectors and class information.
Main Results:
- GammaGAN demonstrated enhanced video quality compared to previous models.
- Achieved relative improvements of 1.61% in PSNR, 1.66% in SSIM, and 0.36% in LPIPS on the MUG facial expression dataset.
- The normalization technique effectively balanced the influence of feature vectors and class embeddings.
Conclusions:
- The proposed GammaGAN model significantly advances conditional video generation.
- The method shows promise for future research in generating high-quality videos from class-conditioned inputs.
- Effective class label integration is crucial for plausible video synthesis.

