Chain-of-Modality: Learning Manipulation Programs from Multimodal Human Videos with Vision-Language-Models
This work addresses the challenge of teaching robots complex manipulation tasks from human demonstrations, though it is incremental by building on existing vision-language models with a new prompting strategy.
The paper tackles the problem of learning manipulation tasks from human videos by incorporating multimodal data (muscle or audio signals) to capture control parameters like force, which visual data alone cannot, resulting in a threefold improvement in accuracy for extracting task plans and control parameters compared to baselines.
Learning to perform manipulation tasks from human videos is a promising approach for teaching robots. However, many manipulation tasks require changing control parameters during task execution, such as force, which visual data alone cannot capture. In this work, we leverage sensing devices such as armbands that measure human muscle activities and microphones that record sound, to capture the details in the human manipulation process, and enable robots to extract task plans and control parameters to perform the same task. To achieve this, we introduce Chain-of-Modality (CoM), a prompting strategy that enables Vision Language Models to reason about multimodal human demonstration data -- videos coupled with muscle or audio signals. By progressively integrating information from each modality, CoM refines a task plan and generates detailed control parameters, enabling robots to perform manipulation tasks based on a single multimodal human video prompt. Our experiments show that CoM delivers a threefold improvement in accuracy for extracting task plans and control parameters compared to baselines, with strong generalization to new task setups and objects in real-world robot experiments. Videos and code are available at https://chain-of-modality.github.io