<div align="center">
<img src="docs/adel-logo.svg" width="160" >
</div>
# AdelaiDet
AdelaiDet is an open source toolbox for multiple instance-level recognition tasks on top of [Detectron2](https://github.com/facebookresearch/detectron2).
All instance-level recognition works from our group are open-sourced here.
To date, AdelaiDet implements the following algorithms:
* [FCOS](configs/FCOS-Detection/README.md)
* [BlendMask](configs/BlendMask/README.md)
* [MEInst](configs/MEInst-InstanceSegmentation/README.md)
* [ABCNet](configs/BAText/README.md)
* [ABCNetv2](configs/BAText#quick-start-abcnetv2)
* [CondInst](configs/CondInst/README.md)
* [SOLO](https://arxiv.org/abs/1912.04488) ([mmdet version](https://github.com/WXinlong/SOLO))
* [SOLOv2](configs/SOLOv2/README.md)
* [BoxInst](configs/BoxInst/README.md) ([video demo](https://www.youtube.com/watch?v=NuF8NAYf5L8))
* [DenseCL](configs/DenseCL/README.md)
* [FCPose](configs/FCPose/README.md)
* [DirectPose](https://arxiv.org/abs/1911.07451) _to be released_
## Models
### COCO Object Detecton Baselines with [FCOS](https://arxiv.org/abs/1904.01355)
Name | inf. time | box AP | download
--- |:---:|:---:|:---
[FCOS_R_50_1x](configs/FCOS-Detection/R_50_1x.yaml) | 16 FPS | 38.7 | [model](https://cloudstor.aarnet.edu.au/plus/s/glqFc13cCoEyHYy/download)
[FCOS_MS_R_101_2x](configs/FCOS-Detection/MS_R_101_2x.yaml) | 12 FPS | 43.1 | [model](https://cloudstor.aarnet.edu.au/plus/s/M3UOT6JcyHy2QW1/download)
[FCOS_MS_X_101_32x8d_2x](configs/FCOS-Detection/MS_X_101_32x8d_2x.yaml) | 6.6 FPS | 43.9 | [model](https://cloudstor.aarnet.edu.au/plus/s/R7H00WeWKZG45pP/download)
[FCOS_MS_X_101_32x8d_dcnv2_2x](configs/FCOS-Detection/MS_X_101_32x8d_2x_dcnv2.yaml) | 4.6 FPS | 46.6 | [model](https://cloudstor.aarnet.edu.au/plus/s/TDsnYK8OXDTrafF/download)
[FCOS_RT_MS_DLA_34_4x_shtw](configs/FCOS-Detection/FCOS_RT/MS_DLA_34_4x_syncbn_shared_towers.yaml) | 52 FPS | 39.1 | [model](https://cloudstor.aarnet.edu.au/plus/s/4vc3XwQezyhNvnB/download)
More models can be found in FCOS [README.md](configs/FCOS-Detection/README.md).
### COCO Instance Segmentation Baselines with [BlendMask](https://arxiv.org/abs/2001.00309)
Model | Name |inf. time | box AP | mask AP | download
--- |:---:|:---:|:---:|:---:|:---:
Mask R-CNN | [R_101_3x](https://github.com/facebookresearch/detectron2/blob/master/configs/COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x.yaml) | 10 FPS | 42.9 | 38.6 |
BlendMask | [R_101_3x](configs/BlendMask/R_101_3x.yaml) | 11 FPS | 44.8 | 39.5 | [model](https://cloudstor.aarnet.edu.au/plus/s/e4fXrliAcMtyEBy/download)
BlendMask | [R_101_dcni3_5x](configs/BlendMask/R_101_dcni3_5x.yaml) | 10 FPS | 46.8 | 41.1 | [model](https://cloudstor.aarnet.edu.au/plus/s/vbnKnQtaGlw8TKv/download)
For more models and information, please refer to BlendMask [README.md](configs/BlendMask/README.md).
### COCO Instance Segmentation Baselines with [MEInst](https://arxiv.org/abs/2003.11712)
Name | inf. time | box AP | mask AP | download
--- |:---:|:---:|:---:|:---:
[MEInst_R_50_3x](https://github.com/aim-uofa/AdelaiDet/configs/MEInst-InstanceSegmentation/MEInst_R_50_3x.yaml) | 12 FPS | 43.6 | 34.5 | [model](https://cloudstor.aarnet.edu.au/plus/s/1ID0DeuI9JsFQoG/download)
For more models and information, please refer to MEInst [README.md](configs/MEInst-InstanceSegmentation/README.md).
### Total_Text results with [ABCNet](configs/BAText/README.md)
Name | inf. time | e2e-hmean | det-hmean | download
--- |:---------:|:---------:|:---------:|:---:
[v1-totaltext](configs/BAText/TotalText/attn_R_50.yaml) | 11 FPS | 67.1 | 86.0 | [model](https://cloudstor.aarnet.edu.au/plus/s/t2EFYGxNpKPUqhc/download)
[v2-totaltext](configs/BAText/TotalText/v2_attn_R_50.yaml) | 7.7 FPS | 71.8 | 87.2 | [model](https://drive.google.com/file/d/1jR5-A-7ITvjdSx3kWVE9bMgh_biMsqcR/view?usp=sharing)
For more models and information, please refer to ABCNet [README.md](configs/BAText/README.md).
### COCO Instance Segmentation Baselines with [CondInst](https://arxiv.org/abs/2003.05664)
Name | inf. time | box AP | mask AP | download
--- |:---:|:---:|:---:|:---:
[CondInst_MS_R_50_1x](configs/CondInst/MS_R_50_1x.yaml) | 14 FPS | 39.7 | 35.7 | [model](https://cloudstor.aarnet.edu.au/plus/s/Trx1r4tLJja7sLT/download)
[CondInst_MS_R_50_BiFPN_3x_sem](configs/CondInst/MS_R_50_BiFPN_3x_sem.yaml) | 13 FPS | 44.7 | 39.4 | [model](https://cloudstor.aarnet.edu.au/plus/s/9cAHjZtdaAGnb2Q/download)
[CondInst_MS_R_101_3x](configs/CondInst/MS_R_101_3x.yaml) | 11 FPS | 43.3 | 38.6 | [model](https://cloudstor.aarnet.edu.au/plus/s/vWLiYm8OnrTSUD2/download)
[CondInst_MS_R_101_BiFPN_3x_sem](configs/CondInst/MS_R_101_BiFPN_3x_sem.yaml) | 10 FPS | 45.7 | 40.2 | [model](https://cloudstor.aarnet.edu.au/plus/s/2p1ashxl54Su8vv/download)
For more models and information, please refer to CondInst [README.md](configs/CondInst/README.md).
Note that:
- Inference time for all projects is measured on a NVIDIA 1080Ti with batch size 1.
- APs are evaluated on COCO2017 val split unless specified.
## Installation
First install Detectron2 following the official guide: [INSTALL.md](https://github.com/facebookresearch/detectron2/blob/master/INSTALL.md).
*Please use Detectron2 with commit id [9eb4831](https://github.com/facebookresearch/detectron2/commit/9eb4831f742ae6a13b8edb61d07b619392fb6543) if you have any issues related to Detectron2.*
Then build AdelaiDet with:
```
git clone https://github.com/aim-uofa/AdelaiDet.git
cd AdelaiDet
python setup.py build develop
```
If you are using docker, a pre-built image can be pulled with:
```
docker pull tianzhi0549/adet:latest
```
Some projects may require special setup, please follow their own `README.md` in [configs](configs).
## Quick Start
### Inference with Pre-trained Models
1. Pick a model and its config file, for example, `fcos_R_50_1x.yaml`.
2. Download the model `wget https://cloudstor.aarnet.edu.au/plus/s/glqFc13cCoEyHYy/download -O fcos_R_50_1x.pth`
3. Run the demo with
```
python demo/demo.py \
--config-file configs/FCOS-Detection/R_50_1x.yaml \
--input input1.jpg input2.jpg \
--opts MODEL.WEIGHTS fcos_R_50_1x.pth
```
### Train Your Own Models
To train a model with "train_net.py", first
setup the corresponding datasets following
[datasets/README.md](https://github.com/facebookresearch/detectron2/blob/master/datasets/README.md),
then run:
```
OMP_NUM_THREADS=1 python tools/train_net.py \
--config-file configs/FCOS-Detection/R_50_1x.yaml \
--num-gpus 8 \
OUTPUT_DIR training_dir/fcos_R_50_1x
```
To evaluate the model after training, run:
```
OMP_NUM_THREADS=1 python tools/train_net.py \
--config-file configs/FCOS-Detection/R_50_1x.yaml \
--eval-only \
--num-gpus 8 \
OUTPUT_DIR training_dir/fcos_R_50_1x \
MODEL.WEIGHTS training_dir/fcos_R_50_1x/model_final.pth
```
Note that:
- The configs are made for 8-GPU training. To train on another number of GPUs, change the `--num-gpus`.
- If you want to measure the inference time, please change `--num-gpus` to 1.
- We set `OMP_NUM_THREADS=1` by default, which achieves the best speed on our machines, please change it as needed.
- This quick start is made for FCOS. If you are using other projects, please check the projects' own `README.md` in [configs](configs).
## Acknowledgements
The authors are grateful to
Nvidia, Huawei Noah's Ark Lab, ByteDance, Adobe who generously donated GPU computing in the past a few years.
## Citing AdelaiDet
If you use this toolbox in your research or wish to refer to the baseline results published here, please use the following BibTeX entries:
```BibTeX
@misc{tian2019adelaidet,
author = {Tian, Zhi and Chen, Hao and Wang, Xinlong and Liu, Yuliang and Shen, Chunhua},
title = {{AdelaiDet}: A Toolbox for Instance-level Recognition Tasks},
howpublished = {\url{https://git.io/adelaidet}},
year = {2019}
}
```
and relevant publications:
```BibTeX
@inproceedings{tian2019fcos,
title = {{FCOS}: Fully Con
没有合适的资源?快使用搜索试试~ 我知道了~
Blendmask-sourceCode源码+内含如何进行具体安装环境与部署
共1797个文件
py:867个
yaml:463个
pyc:199个
需积分: 0 15 下载量 101 浏览量
2023-01-06
23:55:50
上传
评论
收藏 408.58MB 7Z 举报
温馨提示
使用到的有python3.8.12+CUDA11.1+CUDNN-11.1-Windows11-x64-v22000.556Windows 1000.22000.556.0(win11专业版)+pytorch 1.9.0+cu102 CUDA:0 (GeForce GTX 1650, 4096.0MB)+visual studio2019+pycharm 适用于对Detectron感兴趣的小伙伴,以及会安装detectron2的小伙伴。冰墩墩雪容融的数据集目前是丢失了,大家可以自己去准备一个数据集即可,分的类我的话只有两个的,一个是冰墩墩一个是雪容融,大家可以去我的博客里面查看数据集是如何得到的。
资源推荐
资源详情
资源评论
收起资源包目录
Blendmask-sourceCode源码+内含如何进行具体安装环境与部署 (1797个子文件)
events.out.tfevents.1636162859.DESKTOP-2SUSMGO.5276.0 12KB
events.out.tfevents.1636328102.DESKTOP-2SUSMGO.13992.0 518B
events.out.tfevents.1636162210.DESKTOP-2SUSMGO.2712.0 40B
events.out.tfevents.1636326930.DESKTOP-2SUSMGO.8436.0 40B
events.out.tfevents.1636162793.DESKTOP-2SUSMGO.1588.0 40B
events.out.tfevents.1636162723.DESKTOP-2SUSMGO.5636.0 40B
events.out.tfevents.1636162755.DESKTOP-2SUSMGO.7860.0 40B
pkg_helpers.bash 2KB
_mask.c 617KB
maskApi.c 8KB
setup.cfg 931B
setup.cfg 38B
.clang-format 2KB
cocoeval.cpp 20KB
BezierAlign_cpu.cpp 17KB
ROIAlignRotated_cpu.cpp 16KB
gason.cpp 9KB
torchscript_traced_mask_rcnn.cpp 4KB
caffe2_mask_rcnn.cpp 4KB
vision.cpp 3KB
nms_rotated_cpu.cpp 2KB
vision.cpp 2KB
box_iou_rotated_cpu.cpp 1KB
vision.cpp 429B
custom.css 511B
deform_conv_cuda_kernel.cu 43KB
deform_conv_cuda.cu 31KB
DefROIAlign_cuda.cu 15KB
BezierAlign_cuda.cu 15KB
ROIAlignRotated_cuda.cu 14KB
SwapAlign2Nat_cuda.cu 13KB
ml_nms.cu 5KB
nms_rotated_cuda.cu 5KB
box_iou_rotated_cuda.cu 4KB
cuda_version.cu 622B
cuda_version.cu 122B
Dockerfile 2KB
deploy.Dockerfile 2KB
Dockerfile 1KB
pycocotools-2.0.2-py3.8-win-amd64.egg 95KB
pycocotools-2.0.2-py3.7-win-amd64.egg 94KB
_C.cp38-win_amd64.exp 16KB
_C.cp38-win_amd64.exp 16KB
_mask.cp38-win_amd64.exp 773B
_mask.cp37-win_amd64.exp 773B
.flake8 459B
.gitignore 594B
.gitignore 556B
.gitignore 7B
.gitignore 6B
.gitignore 5B
box_iou_rotated_utils.h 11KB
deform_conv.h 8KB
cocoeval.h 3KB
gason.h 3KB
BezierAlign.h 3KB
ROIAlignRotated.h 3KB
DefROIAlign.h 3KB
maskApi.h 2KB
SwapAlign2Nat.h 1KB
nms_rotated.h 1KB
box_iou_rotated.h 988B
ml_nms.h 787B
MANIFEST.in 15B
input1.jpg 137KB
lazyconfig.jpg 64KB
levenshtein.js 2KB
metrics.json 9KB
last_checkpoint 15B
_C.cp38-win_amd64.lib 27KB
_C.cp38-win_amd64.lib 27KB
_mask.cp38-win_amd64.lib 2KB
_mask.cp37-win_amd64.lib 2KB
LICENSE 10KB
LICENSE 1KB
Makefile 634B
Makefile 630B
MODEL_ZOO.md 57KB
DENSEPOSE_IUV.md 30KB
DENSEPOSE_DATASETS.md 19KB
DENSEPOSE_CSE.md 16KB
datasets.md 14KB
README.md 12KB
INSTALL.md 12KB
README.md 11KB
TOOL_APPLY_NET.md 10KB
models.md 9KB
README.md 9KB
augmentation.md 8KB
deployment.md 8KB
benchmarks.md 7KB
README.md 7KB
lazyconfigs.md 7KB
BOOTSTRAPPING_PIPELINE.md 7KB
README.md 6KB
extend.md 6KB
README.md 5KB
data_loading.md 5KB
README.md 5KB
README.md 4KB
共 1797 条
- 1
- 2
- 3
- 4
- 5
- 6
- 18
资源评论
光芒再现dev
- 粉丝: 149
- 资源: 10
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功