Unleashing the Multi-View Fusion Potential: Noise Correction in VLM for Open-Vocabulary 3D Scene Understanding
Published in TMM, 2026
Reducing vision-language feature noise for open-vocabulary 3D understanding.
Xingyilang Yin
Ph.D. Student · Multimodal Understanding, Generation & 3D Vision
- University of Macau & Great Bay University
- Website
- Google Scholar
- CV
- GitHub