img2threejs/img2threejs
Agent skill for Claude Code and Codex that rebuilds the object in a reference image as procedural Three.js TypeScript code through a gated, vision-reviewed, token-efficient pipeline.
Awesome AI Agents › Image Processing & Analysis Agents
LYL1015/JarvisArt
JarvisArt is the code release for a NeurIPS 2025 paper on an intelligent photo retouching agent. The system couples a multimodal large language model with Adobe Lightroom: given an image and a natural language request, the model reasons about the user intent, decides which retouching operations are needed, and emits Lightroom adjustment parameters instead of generating pixels directly. Because the output is a set of editing instructions rather than a rendered image, edits stay non-destructive, resolution independent and inspectable by the photographer. Training follows two stages. A supervised fine-tuning stage teaches the model the chain-of-thought reasoning and the Lightroom parameter vocabulary using MMArt, a dataset of annotated retouching sessions built on PPR10k and released on Hugging Face. A second stage applies GRPO-R, a group relative policy optimisation variant with retouching specific rewards, to improve the quality and consistency of the produced parameter sets. The repository ships training scripts for both stages, evaluation code, and MMArt-Bench, an evaluation set for measuring retouching agents. A local client implements an Agent-to-Lightroom Protocol so that other AI agents can invoke the retouching pipeline against a running Lightroom Classic installation through LrC scripts; data construction scripts for building further training data are also included. Model weights, the MMArt-PPR10k dataset and a Hugging Face Space demo are published alongside the code, and a follow-up project, JarvisEvo, extends the idea to a self-evolving editor-evaluator loop. The work comes from Xiamen University together with HKUST Guangzhou, CUHK, ByteDance, NUS and Tsinghua. It is aimed at researchers working on multimodal agents and tool use, and at developers who want an agent that operates professional creative software rather than a standalone image generator.
https://github.com/LYL1015/JarvisArt
Also in Image Processing & Analysis Agents
apple/ml-ferret
Ferret is an end-to-end multimodal large language model developed by Apple that excels in fine-grained referring and grounding tasks with open vocabulary, supported by a large-scale dataset and evaluation benchmark for research purposes.
THUDM/CogVLM
CogVLM and CogAgent are state-of-the-art open-source visual language models designed for advanced image understanding, multi-turn dialogue, and GUI agent capabilities, achieving top performance on multiple cross-modal benchmarks.
11cafe/jaaz
Jaaz is a local and free AI design agent that enables users to design, edit, and generate images, posters, and storyboards with advanced AI-powered tools and a creative canvas for fast iterations.
SamurAIGPT/Generative-Media-Skills
Collection of agent skills and workflow recipes giving Claude Code, Cursor, Gemini CLI and OpenCode access to image, video and audio generation models through muapi-cli and an MCP server.
QIN2DIM/hcaptcha-challenger
hCaptcha Challenger is an open-source project that uses multimodal large language models and advanced machine learning techniques to automate solving hCaptcha challenges without relying on third-party services or scripts.
roboflow/inference
Roboflow Inference is a platform that turns any computer or edge device into a command center for deploying and managing computer vision models and workflows, enabling advanced AI-powered visual applications.
mbzuai-oryx/groundingLMM
GLaMM is a groundbreaking multimodal AI model that generates natural language responses integrated with object segmentation masks, enabling advanced visual grounding and conversational tasks.