This workshop has been postponed until autumn 2024. A new announcement will be posted on the BærUt website as soon as all details have been finalised
This workshop provides a technical introduction to deep-learning-based handwritten text recognition (HTR). We will begin with a general introduction to the topic and outline a typical HTR pipeline. Afterwards, each pipeline’s steps will be discussed in detail, including hands-on exercises using PyTorch.
The presented examples will focus on a handwritten document scenario, however the general pipeline and most of the steps are directly applicable to printed material as well. Differences and possible adaptations will be pointed out along the way.
While transformer-based HTR models are gaining popularity, this workshop will focus on the more classical but still state-of-the-art approach of Connectionist Temporal Classification. The workshop will conclude with thoughts on reproducibility and experiences from training HTR models on computing centre resources.
When & Where?
2024-06-11 at University of Oslo
More details and registration for the in-person workshop is available via: https://www.ub.uio.no/english/libraries/dsc/berut/events/workshops/2024-06-11_workshop_2.html
Preliminary Lesson Plan
Please note that the lesson plan may still undergo changes until 1-2 weeks before the workshop.
- Welcome and introductions
- General Introduction to Handwritten Text Recognition
- Data for HTR
- Preprocessing
- HTR Models and the CTC Loss
- Inference and Sequence Decoding
- Performance Metrics
- General Thoughts and Experiences from Implementing HTR Models
- Final questions and wrapping up
Setup Instructions
Installation instructions and related pieces of information will be added below in due time.