co-dino-5scale-9encoder-lsj-r50-3x-coco

Co-DINO (ResNet-50, 3x (36 epochs), COCO)

<!-- Demo banner: 1x2 montage of tools/inference.py video outputs produced with THIS checkpoint. Media lives under assets/ in this repo. The <video> renders on the Hugging Face model page; on GitHub the nested <img> poster shows. --> <p align="center"><video controls autoplay loop muted playsinline width="900" poster="https://huggingface.co/dronefreak/co-dino-5scale-9encoder-lsj-r50-3x-coco/resolve/main/assets/demo_banner_poster.jpg" src="https://huggingface.co/dronefreak/co-dino-5scale-9encoder-lsj-r50-3x-coco/resolve/main/assets/demo_banner.mp4"><img src="https://huggingface.co/dronefreak/co-dino-5scale-9encoder-lsj-r50-3x-coco/resolve/main/assets/demo_banner_poster.jpg" alt="Co-DINO ResNet-50 detections on two dashcam clips" width="900"></video></p> <p align="center"> <a href="https://huggingface.co/tasks/object-detection"><img src="https://img.shields.io/badge/Task-Object_Detection-blue?style=flat-square" alt="Task"></a> <a href="https://github.com/open-mmlab/mmdetection/tree/2.x"><img src="https://img.shields.io/badge/Framework-MMDetection_2.x-0aa1a7?style=flat-square" alt="Framework"></a> <a href="https://github.com/Sense-X/Co-DETR"><img src="https://img.shields.io/badge/Architecture-Co--DETR_(DINO)-purple?style=flat-square" alt="Architecture"></a> <a href="https://arxiv.org/abs/2211.12860"><img src="https://img.shields.io/badge/Venue-ICCV_2023-orange?style=flat-square" alt="Venue"></a> <br> <a href="#evaluation-as-reported-by-the-original-authors"><img src="https://img.shields.io/badge/COCO_val_box_AP-55.4_(reported)-9cf?style=flat-square" alt="COCO val box AP"></a> <a href="#measured-inference-footprint"><img src="https://img.shields.io/badge/Params_(inference)-52.8M-lightgrey?style=flat-square" alt="Params"></a> <a href="co_dino_5scale_9encoder_lsj_r50_3x_coco.py"><img src="https://img.shields.io/badge/Schedule-36_epochs_(3x)-lightgrey?style=flat-square" alt="Schedule"></a> <br> <a href="#license-status"><img src="https://img.shields.io/badge/Weights_license-undetermined-lightgrey?style=flat-square" alt="Weights license"></a> <a href="https://arxiv.org/abs/2211.12860"><img src="https://img.shields.io/badge/Paper-arXiv:2211.12860-B31B1B?style=flat-square" alt="Paper"></a> <a href="https://github.com/dronefreak/Co-DETR"><img src="https://img.shields.io/badge/Code-dronefreak/Co--DETR-black?style=flat-square" alt="Code"></a> </p>

Disclaimer

This is not an official release. All credit for the method, the architecture, and the trained weights belongs entirely to the original authors: Zhuofan Zong, Guanglu Song, and Yu Liu (SenseTime X-Lab). I claim no contribution to the underlying research or training. I host the same checkpoint the authors published to Google Drive, with a model card and a copy of its config file.

Mirror rationale

The original non-ViT Co-DETR checkpoints are distributed through a single shared Google Drive folder, which is inconvenient for automated and reproducible downloads (no per-file addressing, checksums, or versioning). The authors published their ViT-L checkpoints to the Hub but not the ResNet-50 / Swin ones. I mirror this checkpoint to provide stable artifact hosting, an explicit file checksum (see Provenance), and versioned Hub access.

I mirror only the pretrained weights; I do not redistribute any training or evaluation data. I will transfer or remove this mirror at the original authors' request. Please cite the original paper (see Citation) and use the official repository or the maintained fork dronefreak/Co-DETR for code.

---

Quickstart

This is not a plain-PyTorch / transformers model. Inference needs the Co-DETR project code and the OpenMMLab 1.x stack (MMDetection 2.25.3, MMCV-full 1.5.0, PyTorch 1.11). The maintained fork ships a one-command environment setup for exactly this.

# 1. Get the code + a validated environment
git clone https://github.com/dronefreak/Co-DETR && cd Co-DETR
bash tools/setup_codetr_env.sh          # builds the `codetr` conda env
conda activate codetr

# 2. Pull this checkpoint + its config from the Hub
pip install -U "huggingface_hub[cli]"
hf download dronefreak/co-dino-5scale-9encoder-lsj-r50-3x-coco \
    co_dino_5scale_9encoder_lsj_r50_3x_coco.pth co_dino_5scale_9encoder_lsj_r50_3x_coco.py \
    --local-dir checkpoints/

# 3. Run detection on an image / folder / video / webcam
python tools/inference.py \
    --config checkpoints/co_dino_5scale_9encoder_lsj_r50_3x_coco.py \
    --checkpoint checkpoints/co_dino_5scale_9encoder_lsj_r50_3x_coco.pth \
    --input demo/demo.jpg --out-dir outputs/ --save-json

The bundled config is identical to projects/configs/co_dino/co_dino_5scale_9encoder_lsj_r50_3x_coco.py in the repo.

---

What is Co-DETR / this checkpoint

Co-DETR ("DETRs with Collaborative Hybrid Assignments Training") is a training scheme, not a new architecture. It attaches auxiliary heads that use one-to-many label assignment (an ATSS head and a Faster-R-CNN-style RoI head) alongside the DETR decoder's one-to-one Hungarian matching, and feeds the positive proposals those heads produce back into the decoder as extra queries. This makes the encoder's features more discriminative and speeds up convergence. The auxiliary heads are used only during training: this checkpoint's config sets eval_module='detr', so at inference only the Co-DINO query head runs and the RPN / RoI / ATSS heads add no inference-time computation to the deployed detector.

This checkpoint is Co-DINO (5 feature scales, 9 encoder layers, LSJ) with a ResNet-50 backbone:

• Paper: DETRs with Collaborative Hybrid Assignments Training, ICCV 2023
• Config: co_dino_5scale_9encoder_lsj_r50_3x_coco.py (bundled; identical to the Co-DETR repo copy)

---

Intended use & limitations

Intended use. Research and engineering work that needs a COCO-pretrained object detector in the Co-DETR family: as a baseline, a feature/detection backbone, or a starting point for fine-tuning on a custom COCO-format dataset.

Limitations.

• Trained and evaluated on COCO only; predicts the 80 COCO thing categories and nothing else.
• Accuracy can degrade substantially on domain-shifted imagery (aerial, medical, document, non-natural images, etc.).
• The reported 55.4 box AP is the authors' number; I have not reproduced it (see Evaluation).
• No real-time claim. See measured latency for actual timings I got on one GPU.
• Runs only on the legacy OpenMMLab 1.x stack (MMDetection 2.25.3 / MMCV-full 1.5.0 / PyTorch 1.11); no transformers / pipeline() support, and inference: false on the Hub.
• The auxiliary training heads in the checkpoint are not exposed as inference outputs; only the DETR query head's detections are returned.

---

What I checked

Integrity and "does it load and run" checks only. This is not an accuracy reproduction:

• It loads with a full state-dict key match in the codetr env (Python 3.8 / torch 1.11.0+cu113 / mmcv-full 1.5.0 / mmdet 2.25.3): zero missing, zero unexpected keys.
• I ran it end-to-end through tools/inference.py on the two dashcam clips shown in the banner above; it produced valid annotated outputs.
• I did not re-measure COCO AP; the 55.4 box AP below is the authors' number.

Measured inference footprint

Method. I built the model with mmdet.apis.init_detector, then ran inference_detector() on a single image, device='cuda:0', batch size 1, FP32 (no AMP), cudnn.benchmark=False: one warm-up call, then torch.cuda.reset_peak_memory_stats() and 10 timed calls with torch.cuda.synchronize(). max_memory_allocated includes the resident model weights plus activation buffers; it excludes the CUDA context.

CPU inference (--device cpu) also works and needs no GPU; it is slower.

---

Evaluation (as reported by the original authors)

Schedule: 3x (36 epochs), 900 object queries, single-scale test, no TTA. I have not independently re-measured these numbers.

---

Training data

• COCO 2017 train2017: 118k images, 80 things categories. Lin et al., Microsoft COCO: Common Objects in Context, ECCV 2014 (arXiv:1405.0312).
• Backbone initialization: ResNet-50 pretrained on ImageNet-1k (torchvision://resnet50).

I do not redistribute COCO here. Get it from cocodataset.org. This checkpoint was trained using COCO and ImageNet initialization only; no Objects365 pretraining is described for it.

---

License status

The Co-DETR source code is MIT-licensed (© 2022 SenseTime X-Lab, LICENSE). The upstream repository does not provide a separate, explicit license for this checkpoint's weights. The original authors have published other Co-DETR checkpoints (the ViT-L family) on Hugging Face with MIT metadata, but I do not treat that as definitive evidence that these weights are independently licensed under MIT.

Because the applicable rights are undetermined, I set the Hub metadata for this repo to license: unknown. Verify the applicable rights yourself before redistribution or commercial use.

If you are one of the original authors and want this mirror removed or transferred, open an issue on this repo or contact dronefreak and I will action it.

---

Citation

@inproceedings{zong2023detrs,
  title={DETRs with Collaborative Hybrid Assignments Training},
  author={Zong, Zhuofan and Song, Guanglu and Liu, Yu},
  booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
  pages={6748--6758},
  year={2023}
}

The underlying detector and backbone:

@inproceedings{zhu2021deformable,
  title={Deformable DETR: Deformable Transformers for End-to-End Object Detection},
  author={Zhu, Xizhou and Su, Weijie and Lu, Lewei and Li, Bin and Wang, Xiaogang and Dai, Jifeng},
  booktitle={International Conference on Learning Representations (ICLR)},
  year={2021}
}
@inproceedings{he2016deep,
  title={Deep Residual Learning for Image Recognition},
  author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  pages={770--778},
  year={2016}
}

Built with MMDetection / MMCV (OpenMMLab).

---

Acknowledgements

I sincerely thank Zhuofan Zong, Guanglu Song, and Yu Liu for developing Co-DETR and releasing the pretrained weights, and the OpenMMLab team for MMDetection and MMCV. This mirror only makes one of their existing checkpoints easier and more reliable to obtain.

---

Provenance

co_dino_5scale_9encoder_lsj_r50_3x_coco.pth, 828,086,923 bytes, SHA-256 7b527af77fd7fa28711d79c5db362dd6c024c675cc56278c80aadcb1be5d3893. I copied it byte-for-byte and unmodified from the authors' Google Drive folder 1nAXOkzqrEgz-YnXxIEs4d5j9li_kmrnv.

Repo files: co_dino_5scale_9encoder_lsj_r50_3x_coco.pth (weights), co_dino_5scale_9encoder_lsj_r50_3x_coco.py (the MMDetection config needed to build the model), config.json (metadata summary; also what the Hub uses to count downloads), plus assets/demo_banner.mp4 / assets/demo_banner_poster.jpg for the card. config.json is descriptive only, inference: false, and is not a transformers config.