reference-images

中文版 AI 姿势参考图库 · 1439 组 / 2878 张图片 / 完整汉化 · 零依赖静态画廊

本地优先的多模型 AI 图片生成工作台，支持 GPT Image、Gemini、图像编辑、素材库、模板与并发任务。Local-first AI image workbench for GPT Image & Gemini, with editing, reusable assets, templates and task history.

💃 Free CC0 mannequin pose reference library for AI image tools (Nano Banana, ChatGPT, ComfyUI, Midjourney) — 250+ poses × 5 angles in line art, names in ja/zh/en/ko

AI 参考图库中文版｜服装（写实 1,407 套／动画 1,320 套）、脸型 1,499 张、姿势线稿与深度图各 1,439 张、二次元 843 条目。面向 AI 绘画、角色设计与漫剧分镜，支持中文检索、多视角浏览、图片复制与下载。开源静态站点，保留原始来源及使用条件。

GPT Image 2.5 Sunburst & Flare toolkit for agents: generation, reference and masked edits, PNG/JPEG/WebP, dry-runs and validated receipts. Experimental ChatGPT/Codex backend.

本地创作与素材管理工作区：无限画布、PBR 材质库、贴图通道、标签搜索、截图贴图、视频参考、文档与脑图。Windows。

Infinite reference canvas for artists: images, local video and YouTube players on one board, with group boxes and connection lines.

Generate 24, 30 or 40 character reference images with matching LoRA training captions in ComfyUI.

在线看图：https://jasonbour.github.io/pose-costume-gallery-zh/ ｜ 1439 张多角度人体素体线稿 + 1407 套无人衣装图中文索引。零依赖静态站，可直接部署到 GitHub Pages。

Muapi AI Character Consistency API endpoint guide, comparison, and runnable API examples.

3D Reference Tool for Artists

GPT Image 2.5 Candid Photography Direction Skills — natural AI candid photos, friend and partner POV, film and CCD looks, reference control, and 8 visual examples.

Role-separated reference director for AI image generation — composition, lighting, material, and identity anchors as a repeatable Codex skill workflow.

Fast Tauri desktop image reference viewer for gesture drawing practice. Random and sequential slideshow for Linux, Windows, and macOS.

BlockDesigner plugin: Blender-style reference images in the scene. Add a picture, move, rotate and scale it, pick the views it shows in, set its opacity and UV. Listed in Layers as REFERENCE.

Novel framework for Zero-Shot Style Alignment in Text-to-Image generation, using minimal attention sharing and ensuring consistent style transfer without fine-tuning.

A fast, native desktop workspace for organizing reference images. Features drag-and-drop, an infinite canvas, grid views, always-on-top transparent overlays, color picking, text notes, and drawn annotations. Built securely with Rust and Tauri for artists and designers.

English edition of AI Pose Library — browse 1,439 pose groups and 2,878 reference images in a searchable, responsive gallery.

Extended local web UI and API based on NomaDamas/god-tibo-imagen, with reference images, history, and prompt tools.

100 curated Seedance 2.0 anime video prompts with verified input references, original X sources, and example videos.