by America Ortiz | Apr 23, 2025 | Machine Learning
Eliminating Oversaturation and Artifacts of High Guidance Scales in Diffusion ModelsClassifier-free guidance (CFG) improves image quality and prompt alignment in diffusion models, but high guidance scales can cause oversaturation and artifacts. In this paper, we...
by America Ortiz | Apr 23, 2025 | Machine Learning
No Training, No Problem: Rethinking Classifier-Free Guidance for Diffusion Models Classifier-free guidance (CFG) is widely used to improve conditional diffusion models, but it requires special training, and it doesn’t extend naturally to unconditional models. In this...
by America Ortiz | Apr 8, 2025 | Rendering, Video Processing, Visual Computing
Volume Scattering Probability Guiding We demonstrate that direct control over the VSP can significantly improve efficiency and present an unbiased volume rendering algorithm based on an existing resampling framework for precise control over the VSP. October 7, 2024...
by America Ortiz | Feb 27, 2025 | Rendering, Video Processing, Visual Computing
CLIP-Fusion: A Spatio-Temporal Quality Metric for Frame Interpolation In this paper, we aim to leverage semantic feature extraction capabilities of the pre-trained visual backbone of CLIP. Specifically, we adapt its multi-scale approach to our feature extraction...
by America Ortiz | Dec 10, 2024 | Machine Learning
LiteVAE: Lightweight and Efficient Variational Autoencoders for Latent Diffusion Models In this paper, we introduce LiteVAE, a new autoencoder design for LDMs, which leverages the 2D discrete wavelet transform to enhance scalability and computational efficiency over...
