{"id":13631261,"url":"https://github.com/SHI-Labs/OneFormer","last_synced_at":"2025-04-17T21:32:39.737Z","repository":{"id":63162830,"uuid":"561086483","full_name":"SHI-Labs/OneFormer","owner":"SHI-Labs","description":"[CVPR 2023] OneFormer: One Transformer to Rule Universal Image Segmentation","archived":false,"fork":false,"pushed_at":"2024-10-03T03:19:15.000Z","size":8998,"stargazers_count":1586,"open_issues_count":44,"forks_count":135,"subscribers_count":19,"default_branch":"main","last_synced_at":"2025-04-12T09:17:48.990Z","etag":null,"topics":["ade20k","cityscapes","coco","image-segmentation","instance-segmentation","oneformer","panoptic-segmentation","semantic-segmentation","transformer","universal-segmentation"],"latest_commit_sha":null,"homepage":"https://praeclarumjj3.github.io/oneformer","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/SHI-Labs.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}},"created_at":"2022-11-02T23:12:15.000Z","updated_at":"2025-04-09T02:58:03.000Z","dependencies_parsed_at":"2024-01-14T06:53:16.761Z","dependency_job_id":"b3f1d9ab-b24e-438f-a3f0-e2440d770d6e","html_url":"https://github.com/SHI-Labs/OneFormer","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SHI-Labs%2FOneFormer","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SHI-Labs%2FOneFormer/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SHI-Labs%2FOneFormer/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/SHI-Labs%2FOneFormer/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/SHI-Labs","download_url":"https://codeload.github.com/SHI-Labs/OneFormer/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":249381006,"owners_count":21261227,"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","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":["ade20k","cityscapes","coco","image-segmentation","instance-segmentation","oneformer","panoptic-segmentation","semantic-segmentation","transformer","universal-segmentation"],"created_at":"2024-08-01T22:02:18.216Z","updated_at":"2025-04-17T21:32:38.906Z","avatar_url":"https://github.com/SHI-Labs.png","language":"Jupyter Notebook","funding_links":[],"categories":["Jupyter Notebook","Paper List"],"sub_categories":["Seminal Papers"],"readme":"# OneFormer: One Transformer to Rule Universal Image Segmentation\n\n[![Framework: PyTorch](https://img.shields.io/badge/Framework-PyTorch-orange.svg)](https://pytorch.org/) [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/SHI-Labs/OneFormer/blob/main/colab/oneformer_colab.ipynb) [![HuggingFace space](https://img.shields.io/badge/🤗-HuggingFace%20Space-cyan.svg)](https://huggingface.co/spaces/shi-labs/OneFormer) [![HuggingFace transformers](https://img.shields.io/badge/🤗-HuggingFace%20transformers-magenta.svg)](https://huggingface.co/docs/transformers/main/en/model_doc/oneformer) [![YouTube](https://badges.aleen42.com/src/youtube.svg)](https://youtu.be/_Zr1pOi7Chw) [![License](https://img.shields.io/badge/License-MIT-red.svg)](https://opensource.org/licenses/MIT)\n \t\n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/instance-segmentation-on-cityscapes-val)](https://paperswithcode.com/sota/instance-segmentation-on-cityscapes-val?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/instance-segmentation-on-coco-val-panoptic)](https://paperswithcode.com/sota/instance-segmentation-on-coco-val-panoptic?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/panoptic-segmentation-on-ade20k-val)](https://paperswithcode.com/sota/panoptic-segmentation-on-ade20k-val?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/instance-segmentation-on-ade20k-val)](https://paperswithcode.com/sota/instance-segmentation-on-ade20k-val?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/panoptic-segmentation-on-cityscapes-val)](https://paperswithcode.com/sota/panoptic-segmentation-on-cityscapes-val?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/panoptic-segmentation-on-cityscapes-test)](https://paperswithcode.com/sota/panoptic-segmentation-on-cityscapes-test?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/panoptic-segmentation-on-mapillary-val)](https://paperswithcode.com/sota/panoptic-segmentation-on-mapillary-val?p=oneformer-one-transformer-to-rule-universal) \t\n[![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/semantic-segmentation-on-mapillary-val)](https://paperswithcode.com/sota/semantic-segmentation-on-mapillary-val?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/semantic-segmentation-on-coco-1)](https://paperswithcode.com/sota/semantic-segmentation-on-coco-1?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/semantic-segmentation-on-ade20k-val)](https://paperswithcode.com/sota/semantic-segmentation-on-ade20k-val?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/panoptic-segmentation-on-coco-minival)](https://paperswithcode.com/sota/panoptic-segmentation-on-coco-minival?p=oneformer-one-transformer-to-rule-universal) [![PWC](https://img.shields.io/endpoint.svg?url=https://paperswithcode.com/badge/oneformer-one-transformer-to-rule-universal/semantic-segmentation-on-cityscapes-val)](https://paperswithcode.com/sota/semantic-segmentation-on-cityscapes-val?p=oneformer-one-transformer-to-rule-universal)\n\n[Jitesh Jain](https://praeclarumjj3.github.io/), [Jiachen Li](https://chrisjuniorli.github.io/)\u003csup\u003e\u0026dagger;\u003c/sup\u003e, [MangTik Chiu](https://www.linkedin.com/in/mtchiu/)\u003csup\u003e\u0026dagger;\u003c/sup\u003e, [Ali Hassani](https://alihassanijr.com/), [Nikita Orlov](https://www.linkedin.com/in/nukich74/), [Humphrey Shi](https://www.humphreyshi.com/home)\n\n\u003csup\u003e\u0026dagger;\u003c/sup\u003e Equal Contribution\n\n[[`Project Page`](https://praeclarumjj3.github.io/oneformer/)] [[`arXiv`](https://arxiv.org/abs/2211.06220)] [[`pdf`](https://arxiv.org/pdf/2211.06220.pdf)] [[`BibTeX`](#4citation)]\n\nThis repo contains the code for our paper **OneFormer: One Transformer to Rule Universal Image Segmentation**.\n\n\u003cimg src=\"images/teaser.png\" width=\"100%\"/\u003e\n\n#### Features\n\n- OneFormer is the **first** multi-task universal image segmentation framework based on transformers.\n- OneFormer needs to be trained only once with a single universal architecture, a single model, and on a single dataset , to outperform existing frameworks across semantic, instance, and panoptic segmentation tasks.\n- OneFormer uses a task-conditioned joint training strategy, uniformly sampling different ground truth domains (semantic instance, or panoptic) by deriving all labels from panoptic annotations to train its multi-task model.\n- OneFormer uses a task token to condition the model on the task in focus, making our architecture task-guided for training, and task-dynamic for inference, all with a single model.\n\n![OneFormer](images/oneformer.svg)\n\n## Contents\n\n1. [News](#news)\n2. [Installation Instructions](#installation-instructions)\n3. [Dataset Preparation](#dataset-preparation)\n4. [Execution Instructions](#execution-instructions)\n    - [Training](#training)\n    - [Evaluation](#evaluation)\n    - [Demo](#demo)\n5. [Results](#results)\n6. [Citation](#citation)\n\n## News\n\n- **[February 27, 2023]**: OneFormer is accepted to CVPR 2023!\n- **[January 26, 2023]**: OneFormer sets new SOTA performance on the the Mapillary Vistas val (both panoptic \u0026 semantic segmentation) and Cityscapes test (panoptic segmentation) sets. We’ve released the checkpoints too!\n- **[January 19, 2023]**: OneFormer is now available as a part of the 🤗 **HuggingFace [transformers](https://huggingface.co/docs/transformers/main/en/model_doc/oneformer) library** and **[model hub](https://huggingface.co/models?filter=oneformer)**! 🚀\n- **[December 26, 2022]**: Checkpoints for Swin-L OneFormer and DiNAT-L OneFormer trained on ADE20K with 1280\u0026times;1280 resolution released!\n- **[November 23, 2022]**: Roboflow cover OneFormer on [YouTube](https://youtu.be/_Zr1pOi7Chw)! Thanks to [@SkalskiP](https://github.com/SkalskiP) for making the video!\n- **[November 18, 2022]**: Our demo is available on 🤗 [Huggingface Space](https://huggingface.co/spaces/shi-labs/OneFormer)!\n- **[November 10, 2022]**: [**Project Page**](https://praeclarumjj3.github.io/oneformer/), [**ArXiv Preprint**](https://praeclarumjj3.github.io/oneformer/) and [**GitHub Repo**](https://praeclarumjj3.github.io/oneformer/) are public!\n  - OneFormer sets new SOTA on Cityscapes val with single-scale inference on Panoptic Segmentation with **68.5** PQ score and Instance Segmentation with **46.7** AP score!\n  - OneFormer sets new SOTA on ADE20K val on Panoptic Segmentation with **51.5** PQ score and on Instance Segmentation with **37.8** AP!\n  - OneFormer sets new SOTA on COCO val on Panoptic Segmentation with **58.0** PQ score!\n\n## Installation Instructions\n\n- We use Python 3.8, PyTorch 1.10.1 (CUDA 11.3 build).\n- We use Detectron2-v0.6.\n- For complete installation instructions, please see [INSTALL.md](INSTALL.md).\n\n## Dataset Preparation\n\n- We experiment on three major benchmark dataset: ADE20K, Cityscapes and COCO 2017.\n- Please see [Preparing Datasets for OneFormer](datasets/README.md) for complete instructions for preparing the datasets.\n\n## Execution Instructions\n\n### Training\n\n- We train all our models using 8 A6000 (48 GB each) GPUs.\n- We use 8 A100 (80 GB each) for training Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e OneFormer and DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e OneFormer on COCO and all models with ConvNeXt-XL\u003csup\u003e\u0026dagger;\u003c/sup\u003e backbone. We also train the 896x896 models on ADE20K on 8 A100 GPUs.\n- Please see [Getting Started with OneFormer](GETTING_STARTED.md) for training commands.\n\n### Evaluation\n\n- Please see [Getting Started with OneFormer](GETTING_STARTED.md) for evaluation commands.\n\n### Demo\n\n- We provide quick to run demos on Colab [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/SHI-Labs/OneFormer/blob/main/colab/oneformer_colab.ipynb) and Hugging Face Spaces [![Huggingface space](https://img.shields.io/badge/🤗-Huggingface%20Space-cyan.svg)](https://huggingface.co/spaces/shi-labs/OneFormer).\n- Please see [OneFormer Demo](demo/README.md) for command line instructions on running the demo.\n\n## Results\n\n![Results](images/plots.svg)\n\n- \u0026dagger; denotes the backbones were pretrained on ImageNet-22k.\n- Pre-trained models can be downloaded following the instructions given [under tools](tools/README.md/#download-pretrained-weights).\n\n### ADE20K\n\n| Method | Backbone | Crop Size |  PQ   | AP   | mIoU \u003cbr\u003e (s.s) | mIoU \u003cbr\u003e (ms+flip) | #params | config | Checkpoint |\n|   :---:| :---:    |  :---:    | :---: | :---:| :---:           | :---:               | :---:   |  :---: |    :---:   |\n| OneFormer | Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 640\u0026times;640 | 49.8 | 35.9 | 57.0 | 57.7 | 219M | [config](configs/ade20k/swin/oneformer_swin_large_bs16_160k.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/250_16_swin_l_oneformer_ade20k_160k.pth) |\n| OneFormer | Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 896\u0026times;896 | 51.1 | 37.6 | 57.4 | 58.3 | 219M | [config](configs/ade20k/swin/oneformer_swin_large_bs16_160k_896x896.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/896x896_250_16_swin_l_oneformer_ade20k_160k.pth) |\n| OneFormer | Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 1280\u0026times;1280 | 51.4 | 37.8 | 57.0 | 57.7 | 219M | [config](configs/ade20k/swin/oneformer_swin_large_bs16_160k_1280x1280.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/1280x1280_250_16_swin_l_oneformer_ade20k_160k.pth) |\n| OneFormer | ConvNeXt-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 640\u0026times;640 | 50.0 | 36.2 | 56.6 | 57.4 | 220M | [config](configs/ade20k/convnext/oneformer_convnext_large_bs16_160k.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/250_16_convnext_l_oneformer_ade20k_160k.pth) |\n| OneFormer | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 640\u0026times;640 | 50.5 | 36.0 | 58.3 | 58.4 | 223M | [config](configs/ade20k/dinat/oneformer_dinat_large_bs16_160k.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/250_16_dinat_l_oneformer_ade20k_160k.pth) |\n| OneFormer | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 896\u0026times;896 | 51.2 | 36.8 | 58.1 | 58.6 | 223M | [config](configs/ade20k/dinat/oneformer_dinat_large_bs16_160k_896x896.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/896x896_250_16_dinat_l_oneformer_ade20k_160k.pth) |\n| OneFormer | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 1280\u0026times;1280 | 51.5 | 37.1 | 58.3 | 58.7 | 223M | [config](configs/ade20k/dinat/oneformer_dinat_large_bs16_160k_1280x1280.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/1280x1280_250_16_dinat_l_oneformer_ade20k_160k.pth) |\n| OneFormer (COCO-Pretrained) | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 1280\u0026times;1280 | 53.4 | 40.2 | 58.4 | 58.8 | 223M | [config](configs/ade20k/dinat/coco_pretrain_oneformer_dinat_large_bs16_160k_1280x1280_coco_pretrain.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/coco_pretrain_1280x1280_150_16_dinat_l_oneformer_ade20k_160k.pth) \u0026#124; [pretrained](https://shi-labs.com/projects/oneformer/coco/150_16_dinat_l_oneformer_coco_100ep.pth) |\n| OneFormer | ConvNeXt-XL\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 640\u0026times;640 | 50.1 | 36.3 | 57.4 | 58.8 | 372M | [config](configs/ade20k/convnext/oneformer_convnext_xlarge_bs16_160k.yaml) | [model](https://shi-labs.com/projects/oneformer/ade20k/250_16_convnext_xl_oneformer_ade20k_160k.pth) |\n\n### Cityscapes\n\n| Method | Backbone |  PQ   | AP   | mIoU \u003cbr\u003e (s.s) | mIoU \u003cbr\u003e (ms+flip) | #params | config | Checkpoint |\n|   :---:| :---:    | :---: | :---:| :---:      | :---:          | :---:   |  :---: |    :---:   |\n| OneFormer | Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 67.2 | 45.6 | 83.0 | 84.4 | 219M | [config](configs/cityscapes/swin/oneformer_swin_large_bs16_90k.yaml) | [model](https://shi-labs.com/projects/oneformer/cityscapes/250_16_swin_l_oneformer_cityscapes_90k.pth) |\n| OneFormer | ConvNeXt-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 68.5 | 46.5 | 83.0 | 84.0 | 220M | [config](configs/cityscapes/convnext/oneformer_convnext_large_bs16_90k.yaml) | [model](https://shi-labs.com/projects/oneformer/cityscapes/250_16_convnext_l_oneformer_cityscapes_90k.pth) |\n| OneFormer (Mapillary Vistas-Pretrained) | ConvNeXt-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 70.1 | 48.7 | 84.6 | 85.2 | 220M | [config](configs/cityscapes/convnext/mapillary_pretrain_oneformer_convnext_large_bs16_90k.yaml) | [model](https://shi-labs.com/projects/oneformer/cityscapes/mapillary_pretrain_250_16_convnext_l_oneformer_cityscapes_90k.pth) \u0026#124; [pretrained](https://shi-labs.com/projects/oneformer/mapillary/mapillary_pretrain_250_16_convnext_l_oneformer_mapillary_300k.pth) |\n| OneFormer | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 67.6 | 45.6 | 83.1 | 84.0 | 223M | [config](configs/cityscapes/dinat/oneformer_dinat_large_bs16_90k.yaml) | [model](https://shi-labs.com/projects/oneformer/cityscapes/250_16_dinat_l_oneformer_cityscapes_90k.pth) |\n| OneFormer | ConvNeXt-XL\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 68.4 | 46.7 | 83.6 | 84.6 | 372M | [config](configs/cityscapes/convnext/oneformer_convnext_xlarge_bs16_90k.yaml) | [model](https://shi-labs.com/projects/oneformer/cityscapes/250_16_convnext_xl_oneformer_cityscapes_90k.pth) |\n| OneFormer (Mapillary Vistas-Pretrained) | ConvNeXt-XL\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 69.7 | 48.9 | 84.5 | 85.8 | 372M | [config](configs/cityscapes/convnext/mapillary_pretrain_oneformer_convnext_xlarge_bs16_90k.yaml) | [model](https://shi-labs.com/projects/oneformer/cityscapes/mapillary_pretrain_250_16_convnext_xl_oneformer_cityscapes_90k.pth) \u0026#124; [pretrained](https://shi-labs.com/projects/oneformer/mapillary/mapillary_pretrain_250_16_convnext_xl_oneformer_mapillary_300k.pth) |\n\n### COCO\n\n| Method | Backbone |  PQ   |  PQ\u003csup\u003eTh\u003c/sup\u003e   |  PQ\u003csup\u003eSt\u003c/sup\u003e   | AP | mIoU | #params | config | Checkpoint |\n|   :---:| :---:    | :---: | :---:              | :---:              |:---:| :---:| :---:  |  :---: |    :---:   |\n| OneFormer | Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 57.9 | 64.4 | 48.0 | 49.0 | 67.4 | 219M | [config](configs/coco/swin/oneformer_swin_large_bs16_100ep.yaml) | [model](https://shi-labs.com/projects/oneformer/coco/150_16_swin_l_oneformer_coco_100ep.pth) |\n| OneFormer | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 58.0 | 64.3 | 48.4 | 49.2 | 68.1 | 223M | [config](configs/coco/dinat/oneformer_dinat_large_bs16_100ep.yaml) | [model](https://shi-labs.com/projects/oneformer/coco/150_16_dinat_l_oneformer_coco_100ep.pth) |\n\n### Mapillary Vistas\n\n| Method | Backbone |  PQ   | mIoU \u003cbr\u003e (s.s) | mIoU \u003cbr\u003e (ms+flip) | #params | config | Checkpoint |\n|   :---:| :---:    | :---: |:---:            | :---:               | :---:  |  :---: |    :---:   |\n| OneFormer | Swin-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 46.7 | 62.9 | 64.1 | 219M | [config](configs/mapillary_vistas/swin/oneformer_swin_large_bs16_300k.yaml) | [model](https://shi-labs.com/projects/oneformer/mapillary/250_16_swin_l_oneformer_mapillary_300k.pth) |\n| OneFormer | ConvNeXt-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 47.9 | 63.2 | 63.8 | 220M | [config](configs/mapillary_vistas/convnext/oneformer_convnext_large_bs16_300k.yaml) | [model](https://shi-labs.com/projects/oneformer/mapillary/250_16_convnext_l_oneformer_mapillary_300k.pth) |\n| OneFormer | DiNAT-L\u003csup\u003e\u0026dagger;\u003c/sup\u003e | 47.8 | 64.0 | 64.9 | 223M | [config](configs/mapillary_vistas/dinat/oneformer_dinat_large_bs16_300k.yaml) | [model](https://shi-labs.com/projects/oneformer/mapillary/250_16_dinat_l_oneformer_mapillary_300k.pth) |\n\n\n## Citation\n\nIf you found OneFormer useful in your research, please consider starring ⭐ us on GitHub and citing 📚 us in your research!\n\n```bibtex\n@inproceedings{jain2023oneformer,\n      title={{OneFormer: One Transformer to Rule Universal Image Segmentation}},\n      author={Jitesh Jain and Jiachen Li and MangTik Chiu and Ali Hassani and Nikita Orlov and Humphrey Shi},\n      journal={CVPR}, \n      year={2023}\n    }\n```\n\n## Acknowledgement\n\nWe thank the authors of [Mask2Former](https://github.com/facebookresearch/Mask2Former), [GroupViT](https://github.com/NVlabs/GroupViT), and [Neighborhood Attention Transformer](https://github.com/SHI-Labs/Neighborhood-Attention-Transformer) for releasing their helpful codebases.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FSHI-Labs%2FOneFormer","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FSHI-Labs%2FOneFormer","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FSHI-Labs%2FOneFormer/lists"}