Haotian Peng
Papers
1
Total Citations
8
H-Index
1
About
Haotian Peng is a researcher at the forefront of 3D computer vision and robotics, with a primary focus on real-time semantic scene understanding and reconstruction. His most notable contribution is the development of **InstanceFusion**, a pioneering system that achieves real-time, instance-level 3D reconstruction of indoor environments using a single RGBD camera. This work uniquely bridges deep learning with traditional SLAM techniques, enabling the simultaneous detection, segmentation, and reconstruction of individual objects in dynamic scenes. With 8 citations, InstanceFusion has laid critical groundwork for interactive robotics and augmented reality applications, demonstrating how to produce visually compelling, semantically rich 3D models on the fly. Peng’s research addresses the fundamental challenge of moving from raw sensor data to actionable, object-level understanding—a key step toward autonomous systems that can navigate and manipulate their surroundings. His work is particularly valued for its robustness and practicality, offering a scalable solution for real-world deployment. As a rising voice in the field, Peng continues to push the boundaries of how machines perceive and reconstruct complex 3D environments.
Research Focus
Key Achievements
Top Papers
- 1