Seminar CS717: Seminar on Computer Vision (FSS 2025)

The Computer Vision seminar covers recent topics in computer vision. In FSS2025, the seminar will focus on “AI Safety & Robustness”, critically exploring the limitations of machine learning models and discussing defences against model vulnerabilities. This seminar the focus will be on certified robustness.

• This seminar is organized by Prof. Dr.-Ing. Margret Keuper
• Available for Master students (2 SWS, 4 ECTS)
• Prerequisites: solid background in machine learning
• Maximum number of participants is 12 students

In this seminar, you will

• 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

• Register as described below.
• Attend the kickoff meeting at Friday, 21.02.
• Work individually throughout the semester according to the seminar schedule (TBA).
• Meet your advisor for guidance and feedback.
• Flash Presentations: (tentatively in the week of 31.03.-06.04.)
• Final Presentations: (tentatively in the week of 19.05.-25.05.)

Each student works on a topic within the area of the seminar along with an accompanying reference paper. Your presentation and report should explore the topic with an emphasis on the reference paper, but not just the reference paper.

We strongly encourage you to explore the available literature and suggest a topic and reference paper of your own choice. Reference papers should be strong papers from a major venue; contact us if you are unsure.

We provide example topics and reference papers below.

Topic List: 

• Certified Robustness
  • [1] (CERTIFIED!!) ADVERSARIAL ROBUSTNESS FOR FREE!
  • [2] Towards Practical Certifiable Patch Defense with Vision Transformer
  • [3] Certified Adversarial Robustness via Randomized Smoothing
• Fingerprinting / Watermarking
  • [4] Evaluation of Security of ML-based Watermarking: Copy and Removal Attacks
  • [5] Attack-Resilient Image WatermarkingUsing Stable Diffusion
  • [6] Tree-Ring Watermarks: Fingerprints for DiffusionImages that are Invisible and Robust
  • [7] Watermark Stealing in Large Language Models
• Deep Fake Detection
  • [8] Preserving Fairness Generalization in Deepfake Detection
  • [9] UCF: Uncovering Common Features for Generalizable Deepfake Detection
  • [10] DiffusionFake: Enhancing Generalization in DeepfakeDetection via Guided Stable Diffusion
  • [11] FreqBlender: Enhancing DeepFake Detection byBlending Frequency Knowledge
  • [12] Dense Feature Interaction Network for Image Inpainting Localization
• Adversarial Detection
  • [13] Segment and Complete: Defending Object Detectors against Adversarial Patch Attacks with Robust Patch Detection
  • [14] NAPGuard: Towards Detecting Naturalistic Adversarial Patches
  • [15] PatchZero: Defending against Adversarial Patch Attacks by Detecting andZeroing the Patch

You can contact us in case you have a suggestion of your own.

The following survey articles and tutorial are good starting points for getting an overview of the topics of the seminar:

• Concrete Problems in AI Safety, Amodei et al. 2016
• Unsolved Problems in ML Safety, Hendrycks et al. 2022