yfxc / pseudo-3d-tensorflow

Tensorflow implement for Pseudo-3d-residual network.

pseudo-3d-tensorflow

Tensorflow implement for pseudo-3d-residual-network.

---

Author	yfxc
E-mail	[e-mail]
Tensorflow	1.10+（DO NOT SUPPORT 2.0）

---

Introduction

Pseudo-3d-residual-network is mainly used for action recognition,paper url: http://openaccess.thecvf.com/content_ICCV_2017/papers/Qiu_Learning_Spatio-Temporal_Representation_ICCV_2017_paper.pdf

Here is the tensorflow version.

Preparing your own dataset.

Suppose you are about to use UCF dataset.Firstly converting videos to images is necessary. To do this,you could run codes like follows:(Suppose UCF-101 dataset is in the same directory as the code-files.)

• ./process_video2image.sh UCF101
• And next step,you should get the train.list and test.list which you would afterwards fetch from for training data and testing data individually:(number ‘5’ indicates that one-fifth of all data is testing data.)
• ./process_gettxt.sh UCF101 5

Note that:Due to the fact that Relative Path of the video clips exist in 'train.list' and 'test.list', So you must make sure that 'DataGenerator.py' and UCF-101 are in the same directory! or modify the codes by yourself.

Train or Eval model

After getting your own data.You can run python train.py --txt='./train.list' to train model. You can also train and test model in 'tf-p3d-train_eval.ipynb' with jupyter notebook.

Others

• You could change some model settings in 'settings.py',except the options(called 'IS_DA') for whether or not to use data augmentation in 'train.py'.
• Changing the properties for data augmentation in 'DataAugmenter.py'

Updates

• Use tf.layers.batch_normalization(training=...) instead of tf.contrib.layers.batch_norm(is_training=...) which may lead to wrong answers when testing.