| Here are a few projects that are built on detectron2. | |
| They are examples of how to use detectron2 as a library, to make your projects more | |
| maintainable. | |
| ## Projects by Facebook | |
| Note that these are research projects, and therefore may not have the same level | |
| of support or stability as detectron2. | |
| + [DensePose: Dense Human Pose Estimation In The Wild](DensePose) | |
| + [Scale-Aware Trident Networks for Object Detection](TridentNet) | |
| + [TensorMask: A Foundation for Dense Object Segmentation](TensorMask) | |
| + [Mesh R-CNN](https://github.com/facebookresearch/meshrcnn) | |
| + [PointRend: Image Segmentation as Rendering](PointRend) | |
| + [Momentum Contrast for Unsupervised Visual Representation Learning](https://github.com/facebookresearch/moco/tree/master/detection) | |
| + [DETR: End-to-End Object Detection with Transformers](https://github.com/facebookresearch/detr/tree/master/d2) | |
| + [Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation](Panoptic-DeepLab) | |
| + [D2Go (Detectron2Go)](https://github.com/facebookresearch/d2go), an end-to-end production system for training and deployment for mobile platforms. | |
| + [Pointly-Supervised Instance Segmentation](PointSup) | |
| + [Unbiased Teacher for Semi-Supervised Object Detection](https://github.com/facebookresearch/unbiased-teacher) | |
| + [Rethinking "Batch" in BatchNorm](Rethinking-BatchNorm/) | |
| + [Per-Pixel Classification is Not All You Need for Semantic Segmentation](https://github.com/facebookresearch/MaskFormer) | |
| + [Exploring Plain Vision Transformer Backbones for Object Detection](ViTDet/) | |
| + [MViTv2: Improved Multiscale Vision Transformers for Classification and Detection](MViTv2/) | |
| ## External Projects | |
| External projects in the community that use detectron2: | |
| <!-- | |
| - If you want to contribute, note that: | |
| - 1. please add your project to the list and try to use only one line | |
| - 2. the project must provide models trained on standard datasets | |
| Projects are *roughly sorted* by: "score = PaperCitation * 5 + Stars", | |
| where PaperCitation equals the citation count of the paper, if the project is an *official* implementation of the paper. | |
| PaperCitation equals 0 otherwise. | |
| --> | |
| + [AdelaiDet](https://github.com/aim-uofa/adet), a detection toolbox including FCOS, BlendMask, etc. | |
| + [CenterMask](https://github.com/youngwanLEE/centermask2) | |
| + [Res2Net backbones](https://github.com/Res2Net/Res2Net-detectron2) | |
| + [VoVNet backbones](https://github.com/youngwanLEE/vovnet-detectron2) | |
| + [FsDet](https://github.com/ucbdrive/few-shot-object-detection), Few-Shot Object Detection. | |
| + [Sparse R-CNN](https://github.com/PeizeSun/SparseR-CNN) | |
| + [BCNet](https://github.com/lkeab/BCNet), a bilayer decoupling instance segmentation method. | |
| + [DD3D](https://github.com/TRI-ML/dd3d), A fully convolutional 3D detector. | |
| + [detrex](https://github.com/IDEA-Research/detrex), a detection toolbox for transformer-based detection algorithms including Deformable-DETR, DAB-DETR, DN-DETR, DINO, etc. | |