Suche
News
You are here
Winter Term: 2022 Recent Advances in Generative Models
Content
The last few years have seen tremendous advances in the quality of generated images that become more and more photorealisitic. This semester, we are going to discuss such methods including neural radiance fields and diffusion models.
Register for course on moodle: https://moodle.uni-siegen.de/course/view.php?id=30323
Qualification Goals
- Read, understand, and explore scientific literature
- Summarize a current research topic in a concise report (10 single-column pages + references)
- Give two presentations about your topic (3 minutes flash presentation, 15 minutes final presentation)
- Moderate a scientific discussion about the topic of one of your fellow students
- Review a (draft of a) report of a fellow student
Schedule
Kickoff-Meeting: October 14th, 1PM
Flash-Presentations: December 9th (tentative)
Final-Presentation: January 31st/ February 7th (tentative)
Location
H-C 6336/37
Literature
- Point-NeRF: Point-based Neural Radiance Fields
- RegNeRF: Regularizing Neural Radiance Fields for View Synthesis from Sparse Inputs
- Deblur-NeRF: Neural Radiance Fields from Blurry Images
- DIVeR Real-time and Accurate Neural Radiance Fields with Deterministic Integration for Volume Rendering
- Block-NeRF Scalable Large Scene Neural View Synthesis
- Projected GANs Converge Faster
- Denoising Diffusion Probabilistic Models
- PointFlow: 3D Point Cloud Generation with Continuous Normalizing Flows
- Diffusion Probabilistic Models for 3D Point Cloud Generation
Winter Term: 2021 Domain Adaptation and Transfer for deep learning models.
Schedule
Kickoff-Meeting: October 22nd
Flash-Presentations: December 3rd (tentative)
Final-Presentation: January 31st/ February 7th (tentative)
Literatur
- A Fourier-based Framwork for Domain Generalization https://arxiv.org/abs/2105.11120
- FSDR: Frequency Space Domain Randomization for Domain Generalization https://arxiv.org/abs/2103.02370
- Rethinking and Improving the Robustness of Image Style Transfer https://arxiv.org/abs/2104.05623
- Reducing Domain Gap by Reducing Style Bias https://arxiv.org/abs/1910.11645
- What Can Style Transfer and Paintings Do For Model Robustness? https://arxiv.org/abs/2011.14477
- Neural Discrete Representation Learning https://arxiv.org/abs/1711.00937
- Rapid Neural Architecture Search by Learning to Generate Graphs from Datasets https://arxiv.org/abs/2107.00860
- Neural Architecture Transfer https://arxiv.org/abs/2005.05859
- XferNAS: Transfer Neural Architecture Search [1907.08307] XferNAS: Transfer Neural Architecture Search - arXiv
- Transfer NAS: Knowledge Transfer between Search Spaces with Transformer Agents https://arxiv.org/pdf/1906.08102.pdf