Computer Science > Computer Vision and Pattern Recognition
[Submitted on 9 Nov 2023]
Title:Vision Encoder-Decoder Models for AI Coaching
View PDFAbstract:This research paper introduces an innovative AI coaching approach by integrating vision-encoder-decoder models. The feasibility of this method is demonstrated using a Vision Transformer as the encoder and GPT-2 as the decoder, achieving a seamless integration of visual input and textual interaction. Departing from conventional practices of employing distinct models for image recognition and text-based coaching, our integrated architecture directly processes input images, enabling natural question-and-answer dialogues with the AI coach. This unique strategy simplifies model architecture while enhancing the overall user experience in human-AI interactions. We showcase sample results to demonstrate the capability of the model. The results underscore the methodology's potential as a promising paradigm for creating efficient AI coach models in various domains involving visual inputs. Importantly, this potential holds true regardless of the particular visual encoder or text decoder chosen. Additionally, we conducted experiments with different sizes of GPT-2 to assess the impact on AI coach performance, providing valuable insights into the scalability and versatility of our proposed methodology.
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
Connected Papers (What is Connected Papers?)
CORE Recommender (What is CORE?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.