Tag: diffusion-models

wangzhiyaoo/SVFR is a unified framework for video face restoration supporting BFR, colorization, and inpainting tasks in a single diffusion-based pipeline. Built on Stable Video Diffusion, it offers both local inference and Hugging Face demo access.

CatVTON is an ICLR 2025 virtual try-on diffusion model achieving 1024×768 inference under 8GB VRAM through elegant concatenation-based design. With only 49.57M trainable parameters, it democratizes fashion AI for developers and researchers.

Discover See-Through, the SIGGRAPH 2026 research project that automatically decomposes anime illustrations into 23 inpainted, semantically organized layers for 2.5D animation pipelines. Free and open-source.

Netflix open-sourced VOID, an AI model that removes objects from videos with physical interaction awareness. Learn setup, code examples, and advanced techniques for next-generation video editing.

NVIDIA's DreamZero achieves #1 robot benchmark rankings with zero-shot generalization using world action models. Learn how to deploy, fine-tune, and extend this open-source robotics breakthrough.

OmniVoice is an open-source TTS model supporting 600+ languages with zero-shot voice cloning, 40x real-time inference speed, and voice design from text attributes. Built on a novel diffusion language model architecture.

Discover Cool-GenAI-Fashion-Papers, the definitive ACM TIST 2025 repository for generative AI in fashion. With 200+ papers across CVPR, NeurIPS, SIGGRAPH and more, plus fashion-specific metrics and active maintenance, it's the stealth weapon top researchers use to stay ahead.

TransPixeler (CVPR 2025) is the breakthrough AI system that generates real RGBA videos with transparency from text prompts. Learn installation, code examples, and why it's replacing manual rotoscoping.

HunyuanWorld-1.0 is Tencent's open-source framework for generating immersive, explorable 3D worlds from text or images. Learn installation, code examples, and why developers are switching from video-based tools.

MinerU-Diffusion replaces slow autoregressive decoding with parallel block-wise diffusion for document OCR, achieving 3.2x speedup with 99.9% relative accuracy. Complete setup guide, real code examples, and production deployment strategies inside.

PersonaLive is a breakthrough CVPR 2026 diffusion framework that transforms static portraits into infinite, real-time animated streams for live streaming. Learn installation, optimization, and advanced usage with real code examples from the repository.

Discover how Microsoft's TRELLIS.2 transforms 2D images into photorealistic 3D assets using revolutionary O-Voxel sparse structures. Learn installation, code examples, and pro tips in this comprehensive guide.

Discover HY-WorldPlay, Tencent's revolutionary framework for interactive world modeling with real-time latency. Learn how to generate infinite 3D worlds at 24 FPS with perfect geometric consistency using breakthrough memory systems and RL optimization.

LiveAvatar generates real-time audio-driven avatars at 45 FPS with infinite length. This comprehensive guide covers installation, code examples, and advanced usage of Alibaba's breakthrough diffusion model for streaming avatar generation.