DSFD

introduction

This is a tensorflow implement mobilenetv3-ssd framework, which can be easily deployeed on both Android(MNN) and IOS(CoreML) mobile devices.

Purpose: Light detection algorithms that work on mobile devices is widely used, such as face detection, qrcode detection. So there is an easy project contains model training and model converter.

** More convenient, the anchor decode and the postprocess(nms) is included in the model. **

** contact me if u have question [email protected] **

##pretrained model

• baidu disk ( password rj94)
• google drive

fddb
0.955

(mnn inference time 13ms including nms,inputs 480x480, on kirin980

coreml not tested)

requirment

• tensorflow1.14

• tensorpack (for data provider)

• opencv

• python 3.6

• MNNConverter

• tfcoreml

useage

we make face detection as an example:

train

1. download widerface data from http://shuoyang1213.me/WIDERFACE/ and release the WIDER_train, WIDER_val and wider_face_split into ./WIDER, then run python prepare_wider_data.pyit will produce train.txt and val.txt (if u like train u own data, u should prepare the data like this: ...../9_Press_Conference_Press_Conference_9_659.jpg| 483(xmin),195(ymin),735(xmax),543(ymax),1(class) ...... one line for one pic, caution! class should start from 1, 0 means bg)

2. download pretrained model from mbv3-small_minimalistic

3. then, run:

   python train.py

   and if u want to check the data when training, u could set vis in train_config.py as True

4. After training, we should fix some params, such as training_flag, and batch=1, and so on, do as follow:

   4.1 android, mnn

   4.1.1 modify the configs/mbv3_cpnfig.py file 
   
       config.MODEL.continue_train=True
       config.MODEL.pretrained_model='yourmodel.ckpt'
       config.MODEL.deployee='mnn'
   
   4.1.2 python train.py, load and save the ckpt immediately,
   
   4.1.3 python tools/auto_freeze.py
   
   

   4.2 ios, coreml

   4.2.1 modify the configs/mbv3_cpnfig.py file 
   
       config.MODEL.continue_train=True
       config.MODEL.pretrained_model='yourmodel.ckpt'
       config.MODEL.deployee='coreml'
   
   
   4.2.2 python train.py, load and save the ckpt immediately,
   
   4.2.3 python tools/auto_freeze.py
   4.2.4 python tools/convert_to_coreml.py produce coreml_mbv3.mlmodel
   

Outputs of he models converted, outputs are nx4 and nx1, stand for coordinates and scores, and coordinates are normalized

evaluation

default scripts is for face detection. ** fddb **

    python model_eval/fddb.py [--model [TRAINED_MODEL]] [--data_dir [DATA_DIR]]
                          [--split_dir [SPLIT_DIR]] [--result [RESULT_DIR]]
    --model              Path of the saved model,default ./model/detector.pb
    --data_dir           Path of fddb all images
    --split_dir          Path of fddb folds
    --result             Path to save fddb results

example python model_eval/fddb.py --model model/detector.pb --data_dir 'fddb/img/' --split_dir fddb/FDDB-folds/ --result 'result/'

** widerface **

    python model_eval/wider.py [--model [TRAINED_MODEL]] [--data_dir [DATA_DIR]]
                           [--result [RESULT_DIR]]
    --model              Path of the saved model,default ./model/detector.pb
    --data_dir           Path of WIDER
    --result             Path to save WIDERface results

example python model_eval/wider.py --model model/detector.pb --data_dir 'WIDER/WIDER_val/' --result 'result/'

visualization

(caution: i dont know where the demo picture comes from, if u think it's a tort, i would like to delete it.)

if u get a trained model and dont need to work on mobile device, run python tools/auto_freeze.py, it will read the checkpoint file in ./model, and produce detector.pb, then

python vis.py

u can check th code in vis.py to make it runable, it's simple.

TODO:

• Android project.