tensorflow
Training Part
train.py: The primary script for training models.losses.py: Contains definitions for loss functions.models.py: Contains the base class for defining a model, visual similar network model.readers.py: Contains definitions for the Video dataset and Frame dataset readers.
config.py: Configrationenv_utils.py: Functions for identifying different environments.utils.py: Common functions for training.logger.py: logging
- LCVR paper: J. Lee; S. Abu-El-Haija. Large-Scale Content-Only Video Recommendation. ICCV 2017.
- CDML for video understanding paper: J. Lee, S. Abu-El-Haija, B. Varadarajan, A. Natsev. Collaborative Deep Metric Learning for Video Understanding. KDD 2018.
- Youtube-8m dataset homepage: https://research.google.com/youtube8m/
- Youtube-8m paper: S. Abu-El-Haija, N. Kothari, J. Lee, P. Natsev. Youtube-8m: A large-scale video classification benchmark. 27 Sep 2016.
- Youtube-8m starter code: https://github.com/google/youtube-8m
- Triplet loss blog: Triplet Loss and Online Triplet Mining in TensorFlow
- Triplet loss paper: A. Hermans, L. Beyer, B. Leibe. In Defense of the Triplet Loss for Person Re-Identification. arXiv:1703.07737v4
- Triplet loss project in TensorFlow : https://github.com/omoindrot/tensorflow-triplet-loss