Thomas Mampilly
Papers
1
Total Citations
27
H-Index
1
About
Thomas Mampilly is a researcher whose work lies at the intersection of computer vision and natural language processing, with a particular focus on cross-modal coreference resolution. His most-cited paper, "Utilizing Visual Attention for Cross-Modal Coreference Interpretation" (2005), has garnered 27 citations—a notable achievement for a specialized topic. In this work, Mampilly pioneered methods for linking linguistic references to visual elements by modeling how human attention shifts across scenes, enabling machines to interpret phrases like "the red car" or "the woman in the blue dress" by grounding them in visual context. This contribution was foundational for early multimodal AI systems, influencing subsequent research in visual question answering and human-robot interaction. While his citation count reflects the niche nature of his focus, Mampilly’s work demonstrated the critical role of attention mechanisms in bridging language and vision, predating the widespread adoption of attention-based architectures in deep learning. His research remains relevant for scholars exploring how AI can achieve more human-like understanding by integrating perceptual and linguistic cues.
Research Focus
Key Achievements
Top Papers
- 1Utilizing Visual Attention for Cross-Modal Coreference Interpretation27 citations · 2005