>mfw Resource news

10/10/2025

>DreamOmni2: Multimodal Instruction-based Editing and Generation
https://github.com/dvlab-research/DreamOmni2

>A Large-scale Dataset for Robust Complex Anime Scene Text Detection
https://huggingface.co/datasets/deepghs/AnimeText

>One Stone with Two Birds: A Null-Text-Null Frequency-Aware Diffusion Models for Text-Guided Image Inpainting
https://github.com/htyjers/NTN-Diff

>Rectified-CFG++ for Flow Based Models
https://rectified-cfgpp.github.io

>AI Avatar Factory: Create and manage AI-powered avatars with ease
https://github.com/AA-Factory/aafactory

>Equilibrium Matching: Generative Modeling with Implicit Energy-Based Models
https://github.com/raywang4/EqM

>ComfyUI SeC Nodes: State-of-the-art video object segmentation
https://github.com/9nate-drake/Comfyui-SecNodes

>ComfyUI Hunyuan Image 3.0 custom node
https://github.com/bgreene2/ComfyUI-Hunyuan-Image-3

10/09/2025

>StyleKeeper: Prevent Content Leakage using Negative Visual Query Guidance
https://github.com/naver-ai/StyleKeeper

>ComfyUI-Hunyuan-Image-3
https://registry.comfy.org/nodes/ComfyUI-Hunyuan-Image-3

>Pixel-Perfect Depth with Semantics-Prompted Diffusion Transformers
https://pixel-perfect-depth.github.io

>GenPilot: A Multi-Agent System for Test-Time Prompt Optimization in Image Generation
https://github.com/27yw/GenPilot

>Lumina-DiMOO: An Omni Diffusion Large Language Model for Multi-Modal Generation and Understanding
https://synbol.github.io/Lumina-DiMOO

>ChatGPT image snares suspect in deadly Pacific Palisades fire
https://www.bbc.com/news/articles/c8exz5yg14ko

>Pocket Comfy V2.0
https://github.com/PastLifeDreamer/Pocket-Comfy/releases/tag/V2.0

>DC Comics won’t support generative AI: ‘not now, not ever’
https://www.theverge.com/news/797540/dc-comics-jim-lee-no-generative-ai-pledge

10/08/2025

>Continual Learning for Image Captioning through Improved Image-Text Alignment
https://github.com/Gepardius/Taetz_Bordelius_Continual_ImageCaptioning