SqueezeDet for gesture

Forked from SqueezeDet and modified for gesture images in Pascol VOC format. Original paper: SqueezeDet: Unified, Small, Low Power Fully Convolutional Neural Networks for Real-Time Object Detection for Autonomous Driving. Bichen Wu, Forrest Iandola, Peter H. Jin, Kurt Keutzer (UC Berkeley & DeepScale)

CNN pretrained Model:

• Download SqueezeDet model parameters from here, and change to pkl format:

  To do

Training/Validation:

• Prepare the Dataset in Psacol Voc style:

  $Path_to_data/VOC2007/                  |->Annotations/                  |     |-> 00****.xml
                          L-> ...    
                    |->JPEGImages/                  |     L-> 00****.jpg
                    L->ImageSets/Main                        |-> test.txt                        |-> train.txt                        |-> trainval.txt    
                          L-> val.txt

• Now we can start training. Training script can be found in $SQDT_ROOT/scripts/train.sh

  cd $SQDT_ROOT/
  ./scripts/train.sh squeezeDet

• At the same time, you can launch evaluation by

  cd $SQDT_ROOT/
  ./scripts/eval_train.sh
  ./scripts/eval_val.sh

• Finally, to monitor training and evaluation process, you can use tensorboard by

  tensorboard --logdir=$LOG_DIR

Demo