11

Mar

Teaser image to Enlarging the Capability of Diffusion Models for Inverse Problems by Guidance

Enlarging the Capability of Diffusion Models for Inverse Problems by Guidance

Prof. Dr. Jong Chul Ye, Graduate School of Artificial Intelligence, KAIST, Korea

   11.03.2024

   6:15 pm - 7:45 pm

   LMU Munich, Theresienstraße 39, Room B 006

The talk introduces two key innovations for enhancing diffusion models in solving inverse problems:

First, a method leveraging two perpendicular 2D diffusion models to tackle 3D problems by representing 3D data as intersecting 2D slices, addressing dimensionality challenges.

Second, a novel solver that uses text prompts to clarify ambiguities, inspired by human perception, dynamically incorporating textual guidance for more accurate solutions.

Experimental results confirm these approaches significantly improve problem-solving accuracy and reduce ambiguities.


Related

Link to From Machine Learning to Autonomous Intelligence

Munich AI Lectures  •  29.09.2023  •  Plenarssitzungssaal, BAdW

From Machine Learning to Autonomous Intelligence

Join the presentation of Yann LeCun, Chief AI Scientist at Meta AI Research, on how can machines learn like humans.


Link to Thoughts on Artificial Intelligence

Munich AI Lectures  •  14.07.2023  •  LMU Audimax

Thoughts on Artificial Intelligence

Jürgen Schimdhuber, one of leading AI researchers worldwide, shares in his talk in the LMU Audimax his thoughts on AI.


Link to Explainable AI via Semantic Information Pursuit

Munich AI Lectures  •  08.03.2023  •  Livestream on YouTube

Explainable AI via Semantic Information Pursuit

Join the presentation of René Vidal, John Hopkins University.


Link to The highs and lows of performance evaluation: Towards a measurement theory for machine learning

Munich AI Lectures  •  08.02.2023  •  Livestream on YouTube

The highs and lows of performance evaluation: Towards a measurement theory for machine learning

Join the presentation of Peter Flach, University of Bristol.