huangwl18/VoxPoser
VoxPoser uses large language models and vision-language models to zero-shot synthesize trajectories for robotic manipulation tasks.
Not currently ranked — collecting fresh signals.
star history
VoxPoser is a zero-shot method that leverages large language models and vision-language models to compose 3D value maps for robotic manipulation. The system generates robot trajectories from natural language commands without requiring any training data. Implementation is provided in the RLBench simulation environment, demonstrating zero-shot generalization to diverse manipulation tasks.
Frequently asked
- What is huangwl18/VoxPoser?
- VoxPoser uses large language models and vision-language models to zero-shot synthesize trajectories for robotic manipulation tasks.
- Is VoxPoser open source?
- Yes — huangwl18/VoxPoser is open source, released under the MIT license.
- What language is VoxPoser written in?
- huangwl18/VoxPoser is primarily written in Python.
- How popular is VoxPoser?
- huangwl18/VoxPoser has 823 stars on GitHub.
- Where can I find VoxPoser?
- huangwl18/VoxPoser is on GitHub at https://github.com/huangwl18/VoxPoser.