Achuta Kadambi

Papers

Total Citations

130

H-Index

About

Achuta Kadambi is a leading figure in computational imaging and computer vision, whose work bridges the gap between physics-based optics and deep learning. His research centers on exploiting light polarization—the wave-like rotation of light—to solve long-standing problems in visual perception, particularly for transparent and reflective objects that traditionally confound standard cameras and algorithms. Kadambi’s most cited work, “Deep Polarization Cues for Transparent Object Segmentation” (2020, 130 citations), reframes the notoriously difficult task of segmenting textureless, transparent objects by using polarization cues to reveal their shape and boundaries. This breakthrough has profound implications for robotics, autonomous driving, and augmented reality, where reliable object handling is critical. Beyond this, his contributions extend to novel camera designs and physics-aware neural networks that integrate first-principles knowledge into learning systems. Recognized for his innovative approach, Kadambi has received early-career awards and his research is regularly published at top venues like CVPR and Nature journals. His work exemplifies how merging classical optics with modern AI can unlock new capabilities in machine perception.