AI MEETINGS

We organize weekly AI meetings on zoom. We hold these events every Tuesday from 10 am to 11 am unless otherwise noted. We have guests from various universities and industries all over the world. Also, our graduate students present their progress some weeks. If you want to join us in these meetings, please email us at ai-info@ku.edu.tr for the zoom info. 

June 8, 2021

Dr. Fatma Güney, KUIS AI Center, Koç University

 Structure, Motion, and Future Prediction in Dynamic Scenes

In this talk, I’ll talk about what we’ve been working on with Sadra and Kaan* in the last one and a half years in my group**. I’ll start by introducing the view synthesis approach to unsupervised monocular depth and ego-motion estimation by Zhou et al. [1]. I’ll point to its limitation with dynamic objects due to static background assumption and mention a few related works addressing it by conditioning on a given segmentation map. Then, I’ll introduce our approach to jointly reason about segmentation and depth without any conditioning. In the second part, I’ll introduce the stochastic video prediction framework proposed by Denton et al. [2] and show how we extend it to motion space with “SLAMP: Stochastic Latent Appearance and Motion Prediction”. Finally, I’ll talk about how structure and motion from the first part can help stochastic video prediction from the second part in real-world driving scenarios.

[1] T. Zhou, M. Brown, N. Snavely, and D. G. Lowe. Unsupervised learning of depth and ego-motion from video. In Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2017.

[2]  E. Denton and R. Fergus. Stochastic video generation with a learned prior. In Proc. of the International Conf. on Machine learning (ICML), 2018.

*Also in collaboration with Aykut Erdem and Erkut Erdem.

**Work under submission, please do not share.

**Work under submission, please do not share.