Repo AI đang lênCập nhật hằng ngày từ GitHub

[Arxiv 2026] WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory

发现惊艳动效、创意视频与交互艺术，探索背后的源码、提示词与制作过程。Curated motion design, creative video & interactive art — with source code, prompts & making-of resources.

Cosmos-Drive-Dreams: Scalable Synthetic Driving Data Generation with World Foundation Models

Let's finetune video generation models!

Add filters (background blur, etc) to your webcam on Linux.

一句话，做成有封面、有动画的 Live 图。Turn a sentence into a Live Photo with a designed cover and animation.

SLA: Beyond Sparsity in Diffusion Transformers via Fine-Tunable Sparse–Linear Attention

AI-powered video editor SDK built with React. Features canvas timeline, drag-and-drop editing, AI captions, and serverless MP4 export. Perfect for building custom video apps.

[arxiv'26] GAE: Learning a Geometry-Native Latent Space for 3D-Consistent World Generation

Implementation of NÜWA, state of the art attention network for text to video synthesis, in Pytorch

Extension of Wav2Lip repository for processing high-quality videos.

[ECCV 2024] FreeInit: Bridging Initialization Gap in Video Diffusion Models

Video Search SDK for Android Kotlin. Integrate in any video app

Multi-model DAG-driven parallel AI film generation — parallel speedup scales with scene independence; Generate film scenes simultaneously instead of one by one; "把影视生成的执行图从拓扑序变成关键路径最优调度" ; 唯一把场景叙事依赖建模为 DAG、以 CPM 算法驱动并行调度的影视生成引擎

A simple library and set of tools for parsing, modifying, and composing SRT files.

Media Editor (图片编辑、视频编辑)

Generate a video script, voice and a talking face completely with AI

Scene-aware storyboard for DeepSeek Harness: 9-16 numbered cells with timestamps + transcript + an audio envelope, packed into ONE image (384 tokens). Two tools: video_storyboard and video_zoom.

DrawingBotV3 is a software for converting images into vector art

YouTube增强及双语字幕解决方案

本地优先的全栈 AI 创作工作站：AI对话 · AI漫画 · 漫剧创作 · 写作台 · 知识学习五大模块；vLLM / llama.cpp / Transformers 多引擎本地推理，显存智能调度、舱壁隔离自愈、仅供学习使用，欢迎交流指正。（最新进度（除插件系统）受到协议限制，暂时无法提交到仓库）

短剧工作台 | 无限时长视频的 MiniMax H3 工作流 | ComfyUI 本地短剧与长视频生成插件

AI-native video editing built around real creative timelines, generative workflows, and local agent control.

Claude Code Skill that turns any idea into a cinematic, model-ready video prompt — Sora · Kling · Veo · Seedance. 21 genre templates, 5-stage structure, eval-tested. Distilled from the AI short Hollywood director PJ Ace called "one of the best short films I've seen in years."

[ICCV 2025] Light-A-Video: Training-free Video Relighting via Progressive Light Fusion

🎬 AI 视频制作与 AIGC 创作工作流（AI Video Creation Workflow）｜用 AI 制作视频、像素动画与漫画短片｜分镜、素材、配乐、模板与可运行 Python 示例｜Blender 分层动画与剪映制作方法

Vidu S: Real-Time Interactive, Editable, and Spatial Video Generation

[ICCV 2023 & TPAMI 2025] MeViS: A Large-scale Benchmark for Video Segmentation with Motion Expressions

Local Video-LLM powered AI Baby Monitor

Ready-made animated components for Remotion — charts, transitions, text reveals, and more

Native Video editor : Video trim, Audio, Video merge, Slow and fast motion, Text and image, etc...

A blazingly fast programmatic video engine, made for agents.

The free and open source solution for automated video capture and distribution at scale.

Live bilingual subtitles for any app on macOS. Captures audio, transcribes speech, and translates offline — all from your menu bar.

Official Code for DiffMorpher: Unleashing the Capability of Diffusion Models for Image Morphing (CVPR 2024)

Python wrapper for ByteDance's Seedance 2.5 and Seedance 2.5 Flash API — Text-to-Video, Image-to-Video, realistic human faces, native 4K, consistent character generation.

FantasyPortrait: Enhancing Multi-Character Portrait Animation with Expression-Augmented Diffusion Transformers

Automatically synchronize and translate subtitles, or create new ones by transcribing, using pre-trained DNNs, Forced Alignments and Transformers. https://subaligner.readthedocs.io/

[ICLR 2026] OmniWorld: A Multi-Domain and Multi-Modal Dataset for 4D World Modeling

A non-linear editing software that helps you to make nice video.

Video Search and Streaming Agent 🕵️‍♂️

SiamMOT: Siamese Multi-Object Tracking

A flutter package for editing video written in pure Dart with fully customizable UI. Supports crop, trim, rotation and cover selection.

Displays subtitles in .ass format from JavaScript. Supports most SSA/ASS features, easily integrates with HTML5 videos.

Frametell: turn your ideas and material into vivid explainer videos with an AI director agent. Local-first, any model.