Pengyu Yin

I am a 3rd year Ph.D. candidate at EEE, NTU in Singapore, supervised by Prof. Lihua Xie. I obtained both my master (2021) and bacholors (2018) degree at AIAR in XJTU, supervised by Prof. Shaoyi Du. I joined a double degree program from 2016-2018 and got the Diplôme d'Ingénieur from CentraleSupélec.

Email / Google Scholar / Github

Research

I'm interested in pose estimation problems in robotics. Recently I've been especially keen on certifiably pose estimation e.g., algorithms w/ non-degeneracy and observable correctness certificates.
Do drop me an email if you are interested in any of my research topic and want to have a talk.

	MCD: Diverse Large-Scale Multi-Campus Dataset for Robot Perception Thien-Minh Nguyen, Shenghai Yuan, Thien Hoang Nguyen, Pengyu Yin, Haozhi Cao, Lihua Xie, Maciej Wozniak, Patric Jensfelt, Marko Thiel, Justin Ziegenbein, Noel Blunder CVPR, 2024 Website / Github / arXiv Data set for semantic SLAM and segmentation with a focus on NRE (Non-Repetitive Epicyclic) LiDar.
	Outram: One-shot Global Localization via Triangulated Scene Graph and Global Outlier Pruning Pengyu Yin, Haozhi Cao, Thien-Minh Nguyen, Shenghai Yuan, Shuyang Zhang, Kangcheng Liu, Lihua Xie ICRA, 2024 Code / arXiv One-shot LiDAR global localization leveraging substructures of 3D scene graphs.
	An Image Acquisition Scheme for Visual Odometry based on Image Bracketing and Online Attribute Control Shuyang Zhang, Jinhao He, Bohuan Xue, Jin Wu, Pengyu Yin, Jianhao Jiao, Ming Liu ICRA, 2024 An image acquisition scheme with image bracketing patterns for complex illumination scenes.
	MoPA: Multi-Modal Prior Aided Domain Adaptation for 3D Semantic Segmentation Haozhi Cao, Yuecong Xu, Jianfei Yang, Pengyu Yin, Shenghai Yuan, Lihua Xie ICRA, 2024 Code / arXiv Multi-modal Prior Aided (MoPA) domain adaptation to improve the performance of rare objects in point cloud semantic segmentation.
	LIO-GVM: an Accurate, Tightly-Coupled Lidar-Inertial Odometry with Gaussian Voxel Map Xingyu Ji, Shenghai Yuan, Pengyu Yin, Lihua Xie RAL, 2024 Code / arXiv Robust LIO w/ distribution-based representation.
	Multi-Modal Continual Test-Time Adaptation for 3D Semantic Segmentation Haozhi Cao, Yuecong Xu, Jianfei Yang, Pengyu Yin, Shenghai Yuan, Lihua Xie ICCV, 2023 arXiv Adaptive cross-modal attending module to achieve reliable 3D segmentation in a continual test-time manner.
	Segregator: Global Point Cloud Registration with Semantic and Geometric Cues Pengyu Yin, Shenghai Yuan, Haozhi Cao, Xingyu Ji, Shuyang Zhang, Lihua Xie ICRA, 2023 Code / arXiv Combining point features and object level semantics to achive highly efficient outlier correspondence rejection.
	CoBigICP: Robust and Precise Point Set Registration using Correntropy Metrics and Bidirectional Correspondence Pengyu Yin, Di Wang, Shaoyi Du, Shihui Ying, Yue Gao, Nanning Zheng IROS, 2020 Code / arXiv / video Interprets the liainson between NDT and ICP with a special kind of M-Estimator (Correntropy) from a statistical viewpoint.

Source code and template adapted from Jon Barron.