Maqola

Enhanced Facial Realism in Personalized Diffusion Models: A Memory-Optimized DreamBooth Implementation for Consumer Hardware

Sandeep GuptaDepartment of Electronics and Communication Engineering, Poornima College of Engineering, Jaipur 302022, Rajasthan, IndiaKanad RayAmity Cognitive Computing and Brain Informatics Center, Amity University Rajasthan, Jaipur 303002, IndiaShamim KaiserInstitute of Information Technology, Jahangirnagar University, Savar, Dhaka 1342, BangladeshSazzad HossainDepartment of System Management and Information Security, Samarkand State University, Samarkand 140104, UzbekistanJocelyn FaubertFaubert Lab, School of Optometry, University of Montreal, Montreal, QC H3T 1P1, Canada

Algorithmsjournal2026en

ABI

Annotatsiya

Despite significant progress in general-purpose diffusion-based models capable of producing high-quality media, this approach is still too difficult to implement on consumer/gamer hardware. We present here a memory-optimized DreamBooth framework designed for consumer-grade GPUs with 16 GB of VRAM, that allows for end-to-end image personalization and addresses some of the limitations of existing solutions. Our system reduces peak GPU memory from 22 GB (baseline DreamBooth) to 14.2 GB through novel hierarchical memory management, including attention slicing, Variational Autoencoder (VAE) tiling, gradient accumulation, and gradient checkpointing integrated within the Hugging Face Accelerate ecosystem. The framework further incorporates state-of-the-art techniques for preserving facial features and a comprehensive automated quality management system. The result is a complete end-to-end pipeline achieving a peak memory of 14.2 GB, with quantitative performance (LPIPS: 0.139, SSIM: 0.879, identity: 0.852, and FID: 23.1) competitive with methods requiring significantly more hardware resources.

Hali tarjima qilinmagan

Mavzular

Face recognition and analysis Generative Adversarial Networks and Image Synthesis Facial Nerve Paralysis Treatment and Research

Identifikatorlar

DOI: 10.3390/a19040257

Iqtiboslar va manbalar

0 ta iqtibos33 ta foydalanilgan manba

Koʻrsatkichlar — AkademScholar · Tez orada