Arnau Ramisa

About

Arnau Ramisa is a leading researcher in the intersection of computer vision and robotics, with a primary focus on robotic manipulation of deformable objects and autonomous robot navigation. His most significant contributions lie in developing perception algorithms for cloth manipulation, a notoriously challenging problem in robotics due to the high dimensionality and variability of textile objects. Ramisa pioneered the use of depth and appearance features for informed robot grasping of highly wrinkled clothes, introducing novel 3D descriptors like FINDDD that are two orders of magnitude faster than competing approaches while maintaining high accuracy. His work on learning RGB-D descriptors of garment parts has enabled robots to identify optimal grasping points, with his seminal 2012 paper on informed grasping accumulating 131 citations. Beyond cloth manipulation, Ramisa has made substantial contributions to mobile robot localization, developing robust vision-based methods that combine multiple local feature detectors for reliable place recognition. His research on combining invariant features with panoramic vision for autonomous navigation has been widely cited, and his evaluations of SIFT-based object recognition methods for mobile robots have helped establish best practices in the field. With over 380 total citations across his most-cited works, Ramisa's research continues to influence both the robotics and computer vision communities.

Research Focus

Key Achievements

Top Papers

1. 1
   Using depth and appearance features for informed robot grasping of highly wrinkled clothes

   131 citations · 2012
2. 2
   Learning RGB-D descriptors of garment parts for informed robot grasping

   53 citations · 2014
3. 3
   Robust vision-based robot localization using combinations of local feature region detectors

   39 citations · 2009
4. 4
   FINDDD: A fast 3D descriptor to characterize textiles for robot manipulation

   35 citations · 2013
5. 5
   Combining Invariant Features and the ALV Homing Method for Autonomous Robot Navigation Based on Panoramas

   26 citations · 2011
6. 6
   A Tale of Two Object Recognition Methods for Mobile Robots

   22 citations · 2008
7. 7
   Mobile robot localization using panoramic vision and combinations of feature region detectors

   22 citations · 2008
8. 8
   Determining Where to Grasp Cloth Using Depth Information

   21 citations · 2011
9. 9
   Evaluation of the SIFT Object Recognition Method in Mobile Robots

   21 citations · 2009
10. 10
    A 3D descriptor to detect task-oriented grasping points in clothing

    16 citations · 2016

Key Collaborators

Contact & Links