NAS_FPN_Tensorflow

KO | EN

gitlite — search

#python #ai #llm #go #deep-learning #docker #macos #nodejs #java #windows #awesome #awesome-list #openai #open-source

NAS_FPN_Tensorflow

(★ 217)

NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection.

File Explorer

Download Latest Version (.zip)

- - __init__.py
  - convert_data_to_tfrecord.py
  - convert_data_to_tfrecord_coco.py
  - convert_data_to_tfrecord_voc2012.py
  - image_preprocess.py
  - image_preprocess_multi_gpu.py
  - image_preprocess_multi_gpu_aug.py
  - read_tfrecord.py
  - read_tfrecord_multi_gpu.py
  - read_tfrecord_multi_gpu_aug.py
- - __init__.py
  - get_coco_next_batch.py
- - README.md
- __init__.py
- __init__.py
- tools.py
- - __init__.py
  - anchor_utils.py
  - boxes_utils.py
  - coordinate_convert.py
  - draw_box_in_img.py
  - encode_and_decode.py
  - iou.py
  - mask_utils.py
  - nms.py
  - show_box_in_tensor.py
  - tf_ops.py
- - __init__.py
  - cfgs.py
- - __init__.py
  - anchor_target_layer_without_boxweight.py
  - proposal_opr.py
  - proposal_target_layer.py
- - __init__.py
  - exportPb.py
  - test_exportPb.py
  - test_TensorRT.py
- - .gitignore
  - README.md
- - __init__.py
  - coco_dict.py
  - label_dict.py
- - __init__.py
  - losses.py
- - __init__.py
  - build_whole_network.py
  - layer.py
  - mobilenet_v2.py
  - nas_fpn.py
  - resnet.py
  - resnet_gluoncv.py
  - resnet_gluoncv_v1.py
- - __init__.py
  - voc_eval.py
- __init__.py
- setup.py
- - README.md
- - 000144.jpg
  - 000283.jpg
  - 000348.jpg
  - 000449.jpg
  - 000558.jpg
  - 000644.jpg
  - 002832.jpg
  - 002996.jpg
  - 003084.jpg
  - 003502.jpg
  - 003902.jpg
  - 003914.jpg
  - 004029.jpg
- - 000144.jpg
  - 000283.jpg
  - 000348.jpg
  - 000449.jpg
  - 000558.jpg
  - 000644.jpg
  - 002832.jpg
  - 002996.jpg
  - 003084.jpg
  - 003502.jpg
  - 003902.jpg
  - 003914.jpg
  - 004029.jpg
- __init__.py
- cocoval.py
- demo.py
- eval.py
- eval_bdd.py
- eval_coco.py
- eval_coco_pyramid.py
- eval_voc2012.py
- inference.py
- inference_for_coco.py
- multi_gpu_train.py
- multi_gpu_train_aug.py
- multi_gpu_train_warmup_cosine.py
- test.py
- test_coco.py
- test_coco_pyramid.py
- test_pyramid_dota.py
- train.py
- train_for_coco.py
- train_with_placeholder.py
.gitignore
comparison.png
fpn_voc_2007.gif
images.png
LICENSE
README.md
scalars.png

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# NAS-FPN: Learning Scalable Feature Pyramid Architecture for Object Detection

## Abstract     
This repo is based on [FPN](https://github.com/DetectionTeamUCAS/NAS_FPN_Tensorflow), and completed by [YangXue](https://github.com/yangxue0827).

## Train on COCO train2017 and test on COCO val2017 (coco minival).     
![1](https://raw.githubusercontent.com/DetectionTeamUCAS/NAS_FPN_Tensorflow/master/fpn_voc_2007.gif)

## COCO
|Model|Backbone|Train Schedule|GPU|Image/GPU|FP16|Box AP|
|-----|--------|--------------|---|---------|----|---------------|
|Faster (Face++ & Detectron)|R50v1-FPN|1X|8X TITAN Xp|2|no|36.4|
|Faster (SimpleDet)|R50v1-FPN|1X|8X 1080Ti|2|no|36.5|
|Faster (ours)|R50v1-FPN|1X|1X TITAN Xp|1|no|36.1|
|Faster (ours)|R50v1-FPN|1X|4X TITAN Xp|1|no|36.1|

|Model|Backbone|Pyramid method|Train Schedule|GPU|Image/GPU|Stacks|Dimension|3x3 relu|Box AP|
|-----|--------|----|--------------|---|---------|----|----|----|---------------|
|Faster (ours)|R50v1|FPN|1X|4X TITAN Xp|1|0|256|no|36.1|
|Faster (ours)|R50v1|FPN|1X|8X 2080Ti|1|3|256|yes|35.8|
|Faster (ours)|R50v1|NAS-FPN|1X|8X 2080Ti|1|3|256|yes|37.9|
|Faster (ours)|R50v1|NAS-FPN|1X|8X 2080Ti|1|7|256|yes|38.1|
|Faster (ours)|R50v1|NAS-FPN|1X|8X 2080Ti|1|7|384|yes|38.9|

## My Development Environment
1、python3.5 (anaconda recommend)             
2、cuda9.0 **(If you want to use cuda8, please set CUDA9 = False in the cfgs.py file.)**                    
3、[opencv(cv2)](https://pypi.org/project/opencv-python/)    
4、[tfplot](https://github.com/wookayin/tensorflow-plot) (optional)             
5、tensorflow == 1.12

## Download Model
### Pretrain weights
1、Please download [resnet50_v1](http://download.tensorflow.org/models/resnet_v1_50_2016_08_28.tar.gz), [resnet101_v1](http://download.tensorflow.org/models/resnet_v1_101_2016_08_28.tar.gz) pre-trained models on Imagenet, put it to data/pretrained_weights.       
2、**(Recommend)** Or you can choose to use a better backbone, refer to [gluon2TF](https://github.com/yangJirui/gluon2TF).    
* [Baidu Drive](https://pan.baidu.com/s/1GpqKg0dOaaWmwshvv1qWGg), password: 5ht9.          
* [Google Drive](https://drive.google.com/drive/folders/1BM8ffn1WnsRRb5RcuAcyJAHX8NS2M1Gz?usp=sharing)

### Trained weights
**Select a configuration file in the folder ($PATH_ROOT/libs/configs/) and copy its contents into cfgs.py, then download the corresponding [weights](https://github.com/DetectionTeamUCAS/Models/tree/master/NAS_FPN_Tensorflow).**

### Others
1、COCO dataset related      
* [instances_minival2014.json](https://drive.google.com/file/d/1cgyEzdGVfx7zPNUO0lLfm8pu0HfIj3Xv/view?usp=sharing)
* [coco_minival2014.odgt](https://drive.google.com/file/d/1Au55e6lqvuTunNBZO2Cj4Kh9XySyM3ZN/view?usp=sharing)        
* [coco_trainvalmini.odgt](https://drive.google.com/file/d/1TSbPjY7HNDryIL3PB6pkO2EVDAwsyD3k/view?usp=sharing)

## Compile
```  
cd $PATH_ROOT/libs/box_utils/cython_utils
python setup.py build_ext --inplace
```

## Train

1、If you want to train your own data, please note:  
```     
(1) Modify parameters (such as CLASS_NUM, DATASET_NAME, VERSION, etc.) in $PATH_ROOT/libs/configs/cfgs.py
(2) Add category information in $PATH_ROOT/libs/label_name_dict/lable_dict.py     
(3) Add data_name to $PATH_ROOT/data/io/read_tfrecord.py 
```

2、make tfrecord
```  
cd $PATH_ROOT/data/io/  
python convert_data_to_tfrecord_coco.py --VOC_dir='/PATH/TO/JSON/FILE/' 
                                        --save_name='train' 
                                        --dataset='coco'
```

3、multi-gpu train
```  
cd $PATH_ROOT/tools
python multi_gpu_train.py
```

## Eval
```  
cd $PATH_ROOT/tools
python eval_coco.py --eval_data='/PATH/TO/IMAGES/'  
                    --eval_gt='/PATH/TO/TEST/ANNOTATION/'
                    --GPU='0'
```

## Tensorboard
```  
cd $PATH_ROOT/output/summary
tensorboard --logdir=.
``` 
![3](https://raw.githubusercontent.com/DetectionTeamUCAS/NAS_FPN_Tensorflow/master/images.png)
![4](https://raw.githubusercontent.com/DetectionTeamUCAS/NAS_FPN_Tensorflow/master/scalars.png)

## Reference
1、https://github.com/endernewton/tf-faster-rcnn   
2、https://github.com/zengarden/light_head_rcnn   
3、https://github.com/tensorflow/models/tree/master/research/object_detection        
4、https://github.com/CharlesShang/FastMaskRCNN       
5、https://github.com/matterport/Mask_RCNN      
6、https://github.com/msracver/Deformable-ConvNets

NAS_FPN_Tensorflow

File Explorer

Today's picks

Was this content helpful?

Recent Feedback

NAS_FPN_Tensorflow

File Explorer

Today's picks

Was this content helpful?

Recent Feedback

Download README