Oussama Remil

Nanjing University of Aeronautics and Astronautics

Papers

Total Citations

H-Index

About

Oussama Remil is a researcher whose work lies at the intersection of computer vision, 3D scene understanding, and video segmentation. His key contributions focus on developing robust methods for analyzing complex visual data, particularly in indoor environments. Remil’s most cited paper, “Object Detection and Tracking Under Occlusion for Object-Level RGB-D Video Segmentation” (2017, 40 citations), addresses the challenging problem of achieving globally consistent segmentation across long RGB-D video sequences. By integrating object detection and tracking, his work enables more reliable scene understanding and robotic grasping, even under occlusion—a critical advancement for real-world applications. Additionally, his research on “Modeling indoor scenes with repetitions from 3D raw point data” (2017, 15 citations) tackles the difficulty of parsing cluttered 3D point clouds by leveraging repetitive structural patterns common in man-made environments. This work contributes to efficient 3D reconstruction and semantic mapping. Through these contributions, Remil has advanced the state of the art in object-level segmentation and 3D scene modeling, providing foundational techniques for autonomous systems and augmented reality.