:orphan:



.. _sphx_glr_build_examples_torch_action_recognition:

Action Recognition
==================



.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="`Kinetics400 &lt;https://deepmind.com/research/open-source/kinetics&gt;`_  is an action recognition d...">

.. only:: html

 .. figure:: /build/examples_torch_action_recognition/images/thumb/sphx_glr_demo_i3d_kinetics400_thumb.png
     :alt: 1. Getting Started with Pre-trained I3D Models on Kinetcis400

     :ref:`sphx_glr_build_examples_torch_action_recognition_demo_i3d_kinetics400.py`

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /build/examples_torch_action_recognition/demo_i3d_kinetics400

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Fine-tuning is an important way to obtain good video models on your own data when you don&#x27;t hav...">

.. only:: html

 .. figure:: /build/examples_torch_action_recognition/images/thumb/sphx_glr_finetune_custom_thumb.png
     :alt: 2. Fine-tuning SOTA video models on your own dataset

     :ref:`sphx_glr_build_examples_torch_action_recognition_finetune_custom.py`

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /build/examples_torch_action_recognition/finetune_custom

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Feature extraction is a very useful tool when you don&#x27;t have large annotated dataset or don&#x27;t h...">

.. only:: html

 .. figure:: /build/examples_torch_action_recognition/images/thumb/sphx_glr_extract_feat_thumb.png
     :alt: 3. Extracting video features from pre-trained models

     :ref:`sphx_glr_build_examples_torch_action_recognition_extract_feat.py`

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /build/examples_torch_action_recognition/extract_feat

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="It is important to have an idea of how to measure a video model&#x27;s speed, so that you can choose...">

.. only:: html

 .. figure:: /build/examples_torch_action_recognition/images/thumb/sphx_glr_speed_thumb.png
     :alt: 4. Computing FLOPS, latency and fps of a model

     :ref:`sphx_glr_build_examples_torch_action_recognition_speed.py`

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /build/examples_torch_action_recognition/speed

.. raw:: html

    <div class="sphx-glr-thumbcontainer" tooltip="Training deep neural networks on videos is very time consuming. For example, training a state-o...">

.. only:: html

 .. figure:: /build/examples_torch_action_recognition/images/thumb/sphx_glr_ddp_pytorch_thumb.png
     :alt: 5. DistributedDataParallel (DDP) Framework

     :ref:`sphx_glr_build_examples_torch_action_recognition_ddp_pytorch.py`

.. raw:: html

    </div>


.. toctree::
   :hidden:

   /build/examples_torch_action_recognition/ddp_pytorch
.. raw:: html

    <div class="sphx-glr-clear"></div>



.. only :: html

 .. container:: sphx-glr-footer
    :class: sphx-glr-footer-gallery


  .. container:: sphx-glr-download sphx-glr-download-python

    :download:`Download all examples in Python source code: examples_torch_action_recognition_python.zip </build/examples_torch_action_recognition/examples_torch_action_recognition_python.zip>`



  .. container:: sphx-glr-download sphx-glr-download-jupyter

    :download:`Download all examples in Jupyter notebooks: examples_torch_action_recognition_jupyter.zip </build/examples_torch_action_recognition/examples_torch_action_recognition_jupyter.zip>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_