Papers
5
Total Citations
104
H-Index
4
About
Farzad Husain’s research lies at the intersection of computer vision and robotics, with a core focus on enabling autonomous systems to perceive, understand, and interact with complex indoor environments. His work uniquely integrates semantic and geometric reasoning, as demonstrated in his highly cited 2016 paper (51 citations), which pioneered efficient pixelwise scene understanding by fusing semantic labels with 3D geometric features from RGB-D sensors like the Microsoft Kinect. Husain made significant contributions to real-time robotic manipulation, developing a system that tracks and grasps moving objects using a geometric particle filter on the affine group (37 citations), a breakthrough for dynamic, real-world applications. He also advanced 3D point cloud recognition through conditional random fields and introduced novel methods for object discovery by leveraging semantic segmentation priors. His joint segmentation and tracking framework for depth videos during human/robot manipulations further showcases his ability to solve practical, real-time challenges. With over 100 total citations, Husain’s work has directly influenced the development of service robots and autonomous systems that must operate reliably in cluttered, unstructured indoor spaces.
Research Focus
Key Achievements
Top Papers
- 1
- 2Realtime tracking and grasping of a moving object from range video37 citations · 2014
- 3Recognizing Point Clouds Using Conditional Random Fields7 citations · 2014
- 4Semantic segmentation priors for object discovery5 citations · 2016
- 5