{"id":31211532,"url":"https://github.com/cyberagentailab/opencole","last_synced_at":"2025-09-21T05:30:49.383Z","repository":{"id":243760633,"uuid":"798046956","full_name":"CyberAgentAILab/OpenCOLE","owner":"CyberAgentAILab","description":"OpenCOLE: Towards Reproducible Automatic Graphic Design Generation [Inoue+, CVPRW2024 (GDUG)]","archived":false,"fork":false,"pushed_at":"2025-03-12T03:04:15.000Z","size":7052,"stargazers_count":78,"open_issues_count":5,"forks_count":10,"subscribers_count":4,"default_branch":"main","last_synced_at":"2025-09-10T07:42:49.423Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"apache-2.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/CyberAgentAILab.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2024-05-09T01:44:29.000Z","updated_at":"2025-08-12T19:42:57.000Z","dependencies_parsed_at":"2024-06-26T09:46:02.099Z","dependency_job_id":"0efcabdc-36ed-4b04-ab9b-411abd14db30","html_url":"https://github.com/CyberAgentAILab/OpenCOLE","commit_stats":null,"previous_names":["cyberagentailab/opencole"],"tags_count":1,"template":false,"template_full_name":null,"purl":"pkg:github/CyberAgentAILab/OpenCOLE","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CyberAgentAILab%2FOpenCOLE","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CyberAgentAILab%2FOpenCOLE/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CyberAgentAILab%2FOpenCOLE/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CyberAgentAILab%2FOpenCOLE/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/CyberAgentAILab","download_url":"https://codeload.github.com/CyberAgentAILab/OpenCOLE/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/CyberAgentAILab%2FOpenCOLE/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":276195627,"owners_count":25601152,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","status":"online","status_checked_at":"2025-09-21T02:00:07.055Z","response_time":72,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2025-09-21T05:30:47.465Z","updated_at":"2025-09-21T05:30:49.365Z","avatar_url":"https://github.com/CyberAgentAILab.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"\u003cdiv align=\"center\"\u003e\n\u003ch1\u003e \u003ca href=\"https://arxiv.org/abs/2406.08232\"\u003eOpenCOLE: Towards Reproducible Automatic Graphic Design Generation \u003c/a\u003e \u003c/h1\u003e\n\n\u003ch4 align=\"center\"\u003e\n    \u003ca href=\"https://naoto0804.github.io/\"\u003eNaoto Inoue*\u003c/a\u003e\u0026emsp;\n    \u003ca href=\"https://scholar.google.co.jp/citations?user=ekIeOUAAAAAJ\u0026hl=en\"\u003eKento Masui*\u003c/a\u003e\u0026emsp;\n    \u003ca href=\"https://scholar.google.co.jp/citations?user=fdXoV1UAAAAJ\"\u003eWataru Shimoda*\u003c/a\u003e\u0026emsp;\n    \u003ca href=\"https://sites.google.com/view/kyamagu\"\u003eKota Yamaguchi\u003c/a\u003e\u0026emsp; (*: equal contribution)\n    \u003cbr\u003e\n    \u003cbr\u003e\n    CyberAgent\n\u003c/h4\u003e\n\n\u003ch4 align=\"center\"\u003e\n\u003ca href=\"https://sites.google.com/view/gdug-workshop/\"\u003eWorkshop on Graphic Design Understanding and Generation\u003c/a\u003e (at CVPR2024)\n\u003c/h4\u003e\n\n![alt text](figs/main_results.png)\n\n\u003c/div\u003e\n\n# Overview\n\n🤔 Automatic generation of graphic designs has recently received considerable attention.\n\n😦 However, the state-of-the-art approaches are **complex** and rely on **proprietary** datasets, which creates reproducibility barriers.\n\n🔥 In this paper, we propose an open framework for automatic graphic design called OpenCOLE, where we build a modified version of the pioneering [COLE [Jia+, arXiv'23]](https://graphic-design-generation.github.io/) and **train our model exclusively on publicly available datasets**.\n\n🚀 Based on GPT4V evaluations, our model shows promising performance comparable to the original COLE. We release the pipeline and training results to encourage **open development**.\n\n# Setup\n\n## Requirements\n\n- [uv](https://astral.sh/blog/uv)\n- [direnv](https://github.com/direnv/direnv)\n\n## Install\n\n```\npoetry install\n```\n\n## Dataset\nOpenCOLE dataset (v1) is available at [`cyberagent/opencole`](https://huggingface.co/datasets/cyberagent/opencole) in HuggingFace dataset hub.\n\n## Pre-trained models\n- text_to_image: [`cyberagent/opencole-stable-diffusion-xl-base-1.0-finetune`](https://huggingface.co/cyberagent/opencole-stable-diffusion-xl-base-1.0-finetune)\n- typography_lmm: [`cyberagent/opencole-typographylmm-llava-v1.5-7b-lora`](https://huggingface.co/cyberagent/opencole-typographylmm-llava-v1.5-7b-lora)\n\n## Environment variables\n\nSome part requires additional environment variables. We recommend to use [direnv](https://direnv.net/).\nPlease copy the template in [.envrc.example](.envrc.example) and modify it on your own.\n\n```bash\ncp .envrc.example .envrc\n```\n\n\n# Inference\n\nPlease refer to [inference.md](./docs/inference.md).\n\n# Evaluation\n\nWe provide a script for GPT4V-based evaluation on generated images.\n\n```python\nuv run python -m opencole.evaluation.eval_gpt4v --input_dir \u003cINPUT_DIR\u003e --output_path \u003cOUTPUT_PATH\u003e\n```\n\n# Training\n\nPlease refer to [training.md](./docs/training.md).\n\n# Citation\n\nIf you find this code useful for your research, please cite our paper:\n\n```\n@inproceedings{inoue2024opencole,\n  title={{OpenCOLE: Towards Reproducible Automatic Graphic Design Generation}},\n  author={Naoto Inoue and Kento Masui and Wataru Shimoda and Kota Yamaguchi},\n  booktitle={Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},\n  year={2024},\n}\n```\n\n# Acknowledgement\nThis repository has been migrated from the internal repo. Despite the fact that commit logs are not visible, all the contributors have made significant contributions to the repository.\n\n- [@proboscis](https://github.com/proboscis): OpenCOLE dataset construction\n- [@shimoda-uec](https://github.com/shimoda-uec): TypographyLMM\n- [@kyamagu](https://github.com/kyamagu): renderer\n- [@naoto0804](https://github.com/naoto0804): other (bunch of) stuffs\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcyberagentailab%2Fopencole","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fcyberagentailab%2Fopencole","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fcyberagentailab%2Fopencole/lists"}