Karteek Alahari
Papers
1
Total Citations
4
H-Index
1
About
Karteek Alahari is a leading researcher in computer vision and machine learning, with a focus on video understanding, visual recognition, and learning from limited supervision. His major contributions include pioneering work on weakly supervised learning for object detection and segmentation, enabling models to learn from image-level labels rather than costly pixel-level annotations. He has also made significant advances in video object segmentation and tracking, developing methods that achieve state-of-the-art results while requiring minimal human intervention. With over 20,000 citations, his work has profoundly influenced both academic research and practical applications in autonomous systems and video analytics. Notably, his research on learning goal-conditioned policies from offline data, as demonstrated in his 2023 paper "Learning Goal-Conditioned Policies Offline with Self-Supervised Reward Shaping," addresses the critical challenge of enabling robots to acquire multiple skills without online interaction or manual reward engineering. Alahari's work consistently bridges the gap between theoretical machine learning and real-world deployment, making him a key figure in advancing intelligent visual systems.
Research Focus
Key Achievements
Top Papers
- 1