Yixing Lao

I'm a PhD Student at The University of Hong Kong, where I work on 3D reconstruction and perception, advised by Hengshuang Zhao.

Previously at Intel, I worked on Open3D and Deep Learning Compilers.

yixing.lao [at] gmail.com  /  Scholar  /  GitHub

profile photo

News

Research

LiT: Unifying LiDAR "Languages" with LiDAR Translator
Yixing Lao, Tao Tang, Xiaoyang Wu, Peng Chen, Kaicheng Yu, Hengshuang Zhao
NeurIPS, 2024
project / code (coming soon) / paper (coming soon)

A framework for unifying LiDAR data from different domains into a single target "language", enabling efficient zero-shot and unified domain detection capabilities.

Pixel-GS: Density Control with Pixel-aware Gradient for 3D Gaussian Splatting
Zheng Zhang, Wenbo Hu, Yixing Lao, Tong He, Hengshuang Zhao
ECCV, 2024
project / code / paper

Pixel-aware gradients and scaled gradient field can improve 3D Gaussian Splatting density control.

LiDAR-NeRF: Novel LiDAR View Synthesis via Neural Radiance Fields
Tao Tang, Longfei Gao, Guangrun Wang, Yixing Lao, Peng Chen, Hengshuang Zhao, Dayang Hao, Xiaodan Liang, Mathieu Salzmann, Kaicheng Yu
ACM Multimedia, 2024
code / paper / video

LiDAR novel-view synthesis with neural radiance fields, enhancing realism for 3D scene reconstructions.

Objects with Lighting: A Real-World Dataset for Evaluating Reconstruction and Rendering for Object Relighting
Benjamin Ummenhofer, Sanskar Agrawal, Rene Sepulveda, Yixing Lao, Kai Zhang, Tianhang Cheng, Stephan R. Richter, Shenlong Wang, German Ros
3DV, 2024
code / paper

A real-world dataset for evaluating inverse rendering methods in object relighting, allowing for a comprehensive analysis of relighting performance.

CorresNeRF: Image Correspondence Priors for Neural Radiance Fields
Yixing Lao, Xiaogang Xu, Zhipeng Cai, Xihui Liu, Hengshuang Zhao
NeurIPS, 2023
project / code / paper

Utilizing image correspondences as NeRF priors improves novel view synthesis and surface reconstruction.

Point Transformer V2: Grouped Vector Attention and Partition-based Pooling
Xiaoyang Wu, Yixing Lao, Li Jiang, Xihui Liu, Hengshuang Zhao
NeurIPS, 2022
code / paper

Grouped Vector Attention and Partition-based Pooling make transformers effective and efficient for point cloud recognition.

ASH: A Modern Framework for Parallel Spatial Hashing in 3D Perception
Wei Dong, Yixing Lao, Michael Kaess, Vladlen Koltun
TPAMI, 2022
code / paper

A high-performance spatial hashing framework on GPU, demonstrating superior performance with fewer lines of code on various tasks such as volumetric reconstruction, non-rigid registration, and spatially varying appearance refinement.



Software

CamTools: Camera Tools for Computer Vision
project / code

A collection of tools for handling cameras in computer vision. Plot, convert, project, ray cast, and do more with your cameras. A must-have for debugging NeRFs and 3D reconstruction.

Open3D: A Modern Library for 3D Data Processing
project / code

The leading open-source library for 3D processing with 400K+ monthly downloads from PyPI. Open3D exposes a set of carefully selected data structures and algorithms in both C++ and Python.


Check out the source code for this website. Yep, it's yet another Jon Barron site.
Last updated Oct 2024.