As a part of Visual Media Lab. seminars, today I had a presentation on how to generate time-lapse videos from one single static image using deep learning, which is the main part of my master thesis. In my presentation I covered these papers:
- “Learning Temporal Transformations From Time-Lapse Videos” by Yipin Zhou and Tamara L.Berg published in ECCV 2016
- “Generating Videos with Scene Dynamics” by Carl Vondrick, Hamed Pirsiavash, Antonio Torralba published in NeurIPS2016
- “Learning to Generate Time-Lapse Videos Using Multi-Stage Dynamic Generative Adversarial Networks” by Wei Xiong, Wenhan Luo, Lin Ma, Wei Liu, Jiebo Luo published in CVPR2018
Here you can find the my presentation and also the video of it.