Anurag Dixit

Papers

Total Citations

H-Index

About

Anurag Dixit is a computer vision researcher whose work sits at the intersection of deep learning efficiency and scene understanding. His most notable contribution, "Depth Estimation with Simplified Transformer" (2022), addresses a critical challenge in modern computer vision: making powerful transformer-based architectures practical for real-world, latency-sensitive deployment. While transformers have rapidly become the dominant paradigm across vision tasks — from image classification to dense prediction — Dixit recognized that their computational demands pose significant barriers to adoption in edge computing and time-critical applications. His research focuses on streamlining transformer architectures without sacrificing the performance gains that have made them state-of-the-art, specifically applying these efficiency innovations to monocular depth estimation, a foundational task in robotics, autonomous driving, and augmented reality. With 11 citations since its publication, this work has begun attracting attention from the research community interested in bridging the gap between high-performing models and practical deployment constraints. Dixit's contributions reflect a growing and important trend in AI research: ensuring that cutting-edge models are not only accurate, but accessible and deployable in the real world.