Boubakeur Boufama

About

Boubakeur Boufama is a leading researcher in computer vision, with a career-long focus on 3D reconstruction, camera calibration, and visual localization. His foundational work, "Understanding positioning from multiple images" (1995, 19 citations), established key principles for extracting spatial information from uncalibrated camera systems. Boufama pioneered techniques for self-calibration, notably developing methods for stationary non-rotating zooming cameras (2014, 6 citations) and stereo heads mounted on robot arms (1994, 6 citations)—work that enabled precise Euclidean structure recovery without pre-calibrated equipment. His research on "Using stereo geometry towards accurate 3D reconstruction" (2009, 6 citations) addressed how stereo vision geometry impacts reconstruction quality, advancing practical applications in robotics and autonomous systems. More recently, Boufama has explored human activity recognition through nonlinear kernel techniques (2012, 3 citations) and location determination from single training videos (2019, 2 citations). His contributions have been instrumental in bridging theoretical camera geometry with real-world robotic vision, making him a respected figure in the field whose work continues to influence modern computer vision systems.

Research Focus

Key Achievements

Top Papers

1. 1
   Understanding positioning from multiple images

   19 citations · 1995
2. 2
   Self-calibration of stationary non-rotating zooming cameras

   6 citations · 2014
3. 3
   Using stereo geometry towards accurate 3D reconstruction

   6 citations · 2009
4. 4
   Self calibration of a stereo head mounted onto a robot arm

   6 citations · 1994
5. 5
   The Advantage of Mounting a Camera onto a Robot Arm

   4 citations · 1995
6. 6
   Nonlinear-Based Human Activity Recognition Using the Kernel Technique

   3 citations · 2012
7. 7
   Determining Location and Detecting Changes Using a Single Training Video

   2 citations · 2019

Key Collaborators

Contact & Links