Dominick Reilly
Papers
1
Total Citations
4
H-Index
1
About
Dominick Reilly is a rising researcher in computer vision, with a focus on integrating geometric and semantic understanding into visual models. His work centers on pose-aware representation learning, particularly within Vision Transformers, aiming to bridge the gap between raw pixel data and structured spatial reasoning. His most cited paper, "Seeing the Pose in the Pixels," introduces a novel framework for embedding explicit pose information—such as human skeletons or robotic arm configurations—directly into transformer architectures, enabling more robust performance in tasks like human action recognition and robot imitation learning. This contribution addresses a critical limitation of conventional vision models, which often overlook the spatial relationships that define human and robotic motion. With 4 citations since its 2023 publication, Reilly’s work is gaining traction for its potential to advance embodied AI and human-robot interaction. By grounding visual perception in pose-aware representations, he is helping to build machines that not only see, but understand the dynamic, pose-rich world around them.
Research Focus
Key Achievements
Top Papers
- 1