Scholar
Tai-Yu (Daniel) Pan
Google Scholar ID: M1_TnJsAAAAJ
Google
Citations & Impact
Citations
275
H-index
6
i10-index
6
Publications
16
Co-authors
16
list available
Publications
Browse publications on Google Scholar (top-right) ↗
Resume (English only)
Academic Achievements
- - Publications:
- - TYP, CVPR 2025
- - R&B-POP, ICLR 2025
- - GPC, ICLR 2024
- - OPS, CVPR 2023
- - Pre-Training LiDAR-Based 3D Object Detectors Through Colorization, ICLR 2024
- - Towards Open-World Segmentation of Parts, CVPR 2023
- - One Step at a Time: Long-Horizon Vision-and-Language Navigation with Milestones, CVPR 2022
- - Learning with Free Object Segments for Long-Tailed Instance Segmentation, ECCV 2022
- - On Model Calibration for Long-Tailed Object Detection and Instance Segmentation, NeurIPS 2021
- - MosaicOS: A Simple and Effective Use of Object-Centric Images for Long-Tailed Object Detection, ICCV 2021
- - Awards: 2024 OSU Graduate Student Research Award
- - Patents: Filed a patent for sub-object segmentation in November 2024
Research Experience
- - Team lead in SAE AutoDrive Competition II, won second place
- - Involved in multiple research projects such as Transfer Your Perspective, Learning 3D Perception from Others' Predictions, etc.
Education
- - Ph.D. in Computer Science and Engineering, 2024, The Ohio State University, Advisor: Dr. Wei-Lun (Harry) Chao
- - MS in Chemical Engineering / Data Science, 2018, University of Washington
- - BS in Chemical Engineering, 2014, National Taiwan University
Background
- - Research Interests: 3D Perception & Generation, 2D Detection & Segmentation, Vision & Language Models, Imbalanced, Long-Tailed Learning, Representation Learning, Autonomous Driving, Multi-Agent Collaborative Driving
- - Professional Field: Computer Science and Engineering
- - Introduction: Earned a Ph.D. in Computer Science and Engineering from The Ohio State University, supervised by Dr. Wei-Lun (Harry) Chao. His research focuses on large-scale 2D/3D computer vision and machine learning, particularly developing models that learn from imperfect real-world data.
Miscellany
- - Personal Interests: Not mentioned
