My research primarily revolves around Embodied AI, a multidisciplinary domain intersecting computer vision, machine learning, and robotics. Specifically, I am dedicated to exploring how prior knowledge in foundational models can be leveraged to construct general-purpose robots capable of effective generalization in diverse, real-world environments.
We introduce ViLa, a novel approach for long-horizon robotic planning
that leverages GPT-4V to generate a sequence of actionable steps. ViLa empowers robots to
execute complex tasks with a profound understanding of the visual world.
We present Semantic-Geometric Representation (SGR), a universal perception module for robotics that
leverages the rich semantic information of large-scale pre-trained 2D models and inherits the
merits of 3D spatial reasoning.
We conduct the first thorough evaluation of pre-trained vision model performance across different downstream policy learning methods and environments. We discover
that the effectiveness of pre-training is highly dependent on the choice of the downstream policy learning algorithm.
We show that fine-grained features learned with pixel-level self-supervised learning (SSL) objectives are complementary to semantic features from image-level SSL methods. Fusing these features can significantly improve the performance for visual correspondence tasks.