Tej Singh

Papers

Total Citations

154

H-Index

About

Tej Singh is a leading researcher in computer vision, with a focused expertise in human action recognition and video-based activity analysis. His work has significantly advanced the field by systematically evaluating and improving the benchmarks used to train and test action recognition models. Singh’s most influential contribution is his comprehensive 2018 review of human action datasets, which has garnered 84 citations and serves as a foundational resource for researchers seeking to understand the strengths and limitations of popular video benchmarks. He further expanded this analysis in a companion survey (42 citations), offering a critical taxonomy of evaluation methodologies. Beyond benchmarking, Singh has innovated in feature extraction, introducing a visual cognizance-based multi-resolution descriptor that leverages key poses for robust action recognition (28 citations). This work demonstrates his commitment to developing more efficient and accurate representations for complex human movements. Through his systematic reviews and novel algorithmic contributions, Tej Singh has helped shape the standards and tools that drive progress in human activity recognition, making his research essential reading for anyone entering the field.