Santhosh Kumar Ramakrishnan

The University of Texas at Austin, Meta (United States)

Papers

Total Citations

H-Index

About

Santhosh Kumar Ramakrishnan is a researcher at the forefront of embodied AI and egocentric vision, with a focus on enabling machines to understand and interact with the world through human-like perception and memory. His work bridges natural language processing and computer vision, particularly in the domain of long-form video understanding. In his highly influential paper "NaQ: Leveraging Narrations as Queries to Supervise Episodic Memory" (2023, 15 citations), Ramakrishnan introduced a novel paradigm for searching extended egocentric video streams using natural language queries. This work has profound implications for augmented reality and robotics, where an agent must fluidly index past experiences to augment human memory and retrieve relevant information on demand. Additionally, his earlier exploration in "An Exploration of Embodied Visual Exploration" (2021, 7 citations) laid groundwork for how autonomous agents can learn to navigate and understand unfamiliar environments. Ramakrishnan's contributions are shaping how machines can build episodic memories and reason about visual experiences, positioning him as a rising voice in the intersection of embodied AI, human-computer interaction, and cognitive systems.