Research Mission
Build fundamental AI/ML methods for computer vision and language modeling to address societal challenges.
Research Interests
Our research is highly interdisciplinary and collaborative, and interests include large language models, computer vision, text/image/video classification, text recognition (OCR, Handwritten), human action analysis, image quality enhancement and NLP tasks (sentiment analysis, Named entity recognition).
Research Projects
- Vision-based deep learning approach for human fall detection
- Human fall detection on untrimmed videos using large foundational video-understanding model
- Vision-Language Models for human action video understanding and summarization
- Leveraging Generative AI Models for Handwritten Text Image Synthesis
- Trustworthy LLMs
- AttentionHTR: Handwritten Text Recognition Based on Attention Encoder-Decoder Networks
Recent News
- Open Postdoc position: Postdoctoral position in Deep Learning with a focus on Vision-Language Models (Deadline: May 14)
- Ekta Vats’ interview with Beijerstiftelsen: 3 frågor till nya Beijerforskaren Ekta Vats (3 questions for the new Beijer Researcher Ekta Vats)
- Open PhD position: PhD student in Machine Learning with a focus on Vision-Language Models (Deadline: March 28)
- Our paper on Uncovering the Handwritten Text in the Margins: End-to-end Handwritten Text Detection and Recognition is accepted at the 8th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature, co-located with the 18th Conference of the European Chapter of the Association for Computational Linguistic (EACL 2024)
- Jan. 2024: Ola Karrar started her Masters thesis on the topic: Vision-based Deep Learning Approach for Human Fall Detection
- Dec. 2023: Women in Data Science Sweden (WiDS) mentorship program wrap up! Ekta Vats served as a mentor
- Oct. – Dec. 2023: Till Grutschus from Technical University of Munich joined us on an exchange semester. Project: Human fall detection on untrimmed videos using large foundational video-understanding model
- 5 Oct. 2023: Beijerforskardagen
- 4 Oct. 2023: Raphaela M. Heil defended her thesis titled Document Image Processing for Handwritten Text Recognition. Deep Learning-based Transliteration of Astrid Lindgren’s Stenographic Manuscripts. Opponent: Andreas Fischer, University of Applied Sciences and Arts Western Switzerland