Dr. Almut Sophia Koepke

BMBF Junior Research Group

MuMoL: Advancing multi-modal learning from sound, vision, and text for video understanding
Chair of Computer Vision and Artificial Intelligence

Academic Career and Research Areas

Dr. Almut Sophia Koepke studied Mathematics in Freiburg and Oxford. She completed her DPhil in the Visual Geometry Group (VGG) at the University of Oxford. After obtaining her PhD, she was a post-doctoral researcher at the University of Tübingen and a visiting scholar at UC Berkeley (USA). In 2024, she joined TUM to lead a Junior Research Group on Multi-modal Learning.

Dr. Almut Sophia Koepke conducts research in the area of multi-modal learning with sound, vision, and text. With her research group, she aims at advancing multi-modal video understanding.

    Karsten Roth, Jae Myung Kim, A. Sophia Koepke, Oriol Vinyals, Cordelia Schmid, Zeynep Akata: ”Waffling around for performance: Visual classification with random words and broad concepts” ICCV, 2023.

    Otniel-Bogdan Mercea, Lukas Riesch, A. Sophia Koepke, Zeynep Akata: “Audio-visual generalised zero-shot learning with cross-modal attention and language” CVPR, 2022.

    A. Sophia Koepke, Andreea-Maria Oncescu, João F. Henriques, Zeynep Akata, Samuel Albanie: “Audio retrieval with natural language queries: A benchmark study” Transactions on Multimedia, 2022.

    A. Sophia Koepke, Olivia Wiles, Yael Moses, Andrew Zisserman: “Sight to sound: An end-to-end approach for visual piano transcription”. ICASSP, 2020

    Olivia Wiles, A. Sophia Koepke, Andrew Zisserman: “X2Face: A network for controlling face generation by using images, audio, and pose codes”. ECCV, 2018.

    If you wish your profile to be changed or updated please contact TUM Talent Factory.