IDEA 18- Python deep learning inference on video

131 views
Skip to first unread message

Kirtika Malhotra

unread,
Mar 21, 2020, 12:10:10 PM3/21/20
to opencv-gsoc-2020
Greetings, respected mentors

My name is Kirtika Malhotra, and I am a 1st Year MCA (Masters of Computer Applications) student at Indira Gandhi Delhi Technical University for Women, Delhi.

While browsing through the list of Ideas, I found the idea to develop a high-level helper class in python top perform optimized inference on videos with data storage in dataframes for easy access to be an interesting and defiant.

I have previously worked on Deep Learning using python and experimented  OpenCV libraries all along the way.I have used OpenCV along with other deep learning frameworks like TensorFlow and PyTorch. OpenCV has always been a great open-source library for warping images/videos as per our need. This motivates me to contribute to the community and learn from it.

I am currently working on the idea of making a helper class in python to make inferences on Video. 

Github: https://github.com/kirtika-malhotra
LinkedIn: https://www.linkedin.com/in/kirtika-malhotra-a7806a162/

Regards,
Kirtika Malhotra

Ilias Balampanis

unread,
Mar 25, 2020, 7:03:37 AM3/25/20
to opencv-gsoc-2020
Dear Mentors,

my name is Ilias and I'm interested in contributing to you. I checked the ideas list and I would like to implement the 18th idea (Python deep learning inference on video). I also checked that you are the mentor of this project.

Initially, I’m studying at the Technical University of Crete in the Electrical and Software Engineering department and I’m going through the last year. I have completed the course guide and I’m elaborating on my diploma thesis. The main goal of this thesis is to implement a network to train distributedly a Deep Learning model, specifically an RNN model. Here is the GitHub repo of this work. At first glance, you ‘ll see that this project is integrated into C++. However, I would still say that at the beginning of this work I tried to implement it in Python. So, I have enough experience in some basic ML, Math and Data Analysis libraries such as NumPy, Tensorflow, Keras, and Pandas. My work experience and my skills are reported in detail on my website. As concerns, the project requirements, have the appropriate experience on Keras and Tensorflow.

Consequently, I would like to submit a proposal for this task. I will appreciate if you would like to chat with an instant message platform if it isn't an issue for you.

Best regards and thank you,
Ilias Balampanis

Justin Shenk

unread,
Mar 29, 2020, 4:20:18 PM3/29/20
to opencv-gsoc-2020
Hi Ilias,
Thanks for your interest. I am mentor for this project. Your background sounds relevant. Please share any questions you have here so that others may learn. If you have a private question, you may reach me via LinkedIn.

Best regards,
Justin

Justin Shenk

unread,
Mar 29, 2020, 4:22:27 PM3/29/20
to opencv-gsoc-2020
Hi Kirtika,
Thanks for your message. Your background sounds relevant. I am mentoring this project, so if you have any questions feel free to post them and I will try to help. 

Justin Shenk
Reply all
Reply to author
Forward
0 new messages