This and That: Language-Gesture Conditioning of Visual Generative Models for Robotics - Tech Talk #7

Explore a comprehensive tech talk on "This&That" presented by Jeong Joon Park as part of the Lerobot Tech Talk series organized by HuggingFace. Delve into the intersection of machine learning and robotics, covering topics such as visual generative models, language-gesture conditioning, and the application of generated video plans to robot actions. Begin with an introduction and recap of machine learning in robotics, then progress through detailed discussions on visual generative models and their conditioning with language and gestures. Learn how these concepts translate into practical robot actions, and conclude with insights into future developments in the field. Engage with the Q&A session to deepen your understanding of this cutting-edge research. Access additional resources, including the research paper and project page, to further expand your knowledge on this innovative approach to robotics and AI.

- Introduction
- Recap on ML for robotics to date
- Visual generative models
- Language-gesture conditioning of visual generative models
- Generated video plans to robot action
- Conclusion and future work
- Q & A