Ser-Nam Lim

Meta (Israel), University of Central Florida

Papers

Total Citations

H-Index

About

Ser-Nam Lim is a computer vision researcher whose work spans 3D object detection, scene understanding, and motion transfer, with a focus on building unified, scalable frameworks that bridge traditionally siloed domains. His most recognized contribution, "UniMODE: Unified Monocular 3D Object Detection" (2024, 19 citations), addresses one of the field's persistent challenges: developing a single model capable of detecting objects across both indoor and outdoor environments using only monocular camera input. This work is particularly significant for real-world applications such as autonomous navigation and robotics, where versatility across diverse scene geometries is critical. His subsequent follow-up, "Toward Unified 3D Object Detection via Algorithm and Data Unification" (2025), extends this vision further by tackling the algorithmic and data-level heterogeneity that makes multi-domain 3D detection difficult. Additionally, his work on "Self-appearance-aided Differential Evolution for Motion Transfer" (2021) demonstrates breadth beyond detection, contributing to unsupervised image animation research. Lim's research consistently targets unification and generalization — pushing models beyond narrow task-specific boundaries toward robust, real-world applicability, making his contributions increasingly relevant as the field moves toward universal perception systems.