Automatic-Video-Description

Automatic video description developed during internship at Nuance Communication, Montreal,

Using Sequence to Sequence Model (CNN-LSTM) on each frame

Name		Name	Last commit message	Last commit date
Latest commit History 5 Commits
cnn_weights		cnn_weights
coco-caption		coco-caption
result		result
save_model		save_model
.DS_Store		.DS_Store
.gitattributes		.gitattributes
README.md		README.md
cnn_model.py		cnn_model.py
commands.txt		commands.txt
crop_videp.py		crop_videp.py
extractRGBfeatures.py		extractRGBfeatures.py
extract_video_batch.py		extract_video_batch.py
index.tsv		index.tsv
loss.png		loss.png
preprocess_tacos.ipynb		preprocess_tacos.ipynb
rnn_model.py		rnn_model.py
run.py		run.py
tacos.csv		tacos.csv
tacos_dataset.py		tacos_dataset.py
tacos_process.csv		tacos_process.csv
testr.py		testr.py
train.py		train.py

Provide feedback