For some time, human advantage against computer was his or her ability to understand speech, read and intercept visual information. The situation changes as AI applications step by step go closer to human senses. However, there are still challenges, especially in computer vision. While computer vision is able to do accurate image matching with the etalon situation, high variance in possible changes might complicate the automation. Therefore it is very important to adjust the existing solutions, tune up or create new computer vision, language models, which could be adapted to specific manufacturing, robotics case.
The direction for the challenge ideas, which could be converted into multiple final thesis or individual project topics is the following:
- AI for robot visual perception
- AI for human speech perception
- Visual language models for robotics
- Large language models for robotics
The topics were proposed by Tampere University (SustAInLIvWork project member), where the main contact is Roel Pieters. Regarding more details on the challenge and topics contact Pavel Stefanovič, VILNIUS TECH.