Bo Yang

I am an Assistant Professor (2020.11-) in the Department of Computing at The Hong Kong Polytechnic University. I completed my D.Phil degree (2016.10-2020.09) in the Department of Computer Science at University of Oxford, supervised by Profs. Niki Trigoni and Andrew Markham. Prior to Oxford, I obtained an M.Phil degree from The University of Hong Kong and a B.Eng degree from Beijing University of Posts and Telecommunications.

In my D.Phil study, I interned at the Augumented Reality team of Amazon (Palo Alto, CA). In my M.Phil study, I interned at Hong Kong Applied Science and Technology Research Institute. In my undergraduate study, I was an exchange student at Universitat Politècnica de València (Valencia, Spain).

Email / vLAR Group / Google Scholar / Github

I lead the Visual Learning and Reasoning (vLAR) Group, focusing on the fundamental research problems in machine learning, computer vision, and robotics. Our research goal is to build intelligent systems which endow machines to recover, understand, and eventually interact with the real 3D world. This includes accurate and efficient recognition, segmentation and reconstruction of all individual objects within large-scale 3D scenes.

(All emails/CVs are carefully read and evaluated. Only matched candidates will be responded.)

[2023.01.22] Our paper DM-NeRF is accepted by ICLR 2023.

[2023.01.18] Our paper DecoupSkill is accepted by ICRA 2023.

[2022.09.03] Our papers OGC and UnsupObjSeg are accepted by NeurIPS 2022.

[2022.07.03] Our paper SQN is accepted by ECCV 2022.

[2022.05.26] Our extension of SpinNet is accepted by TPAMI.

[2022.05.04] We are going to organize The 2nd Challenge on Point Cloud Understanding at ECCV 2022.

[2022.04.19] Our paper RangeUDF is on arXiv.

[2021.11.11] Our extension of SensatUrban is accepted by IJCV.

[2021.07.23] Our paper GRF for neural rendering is accepted by ICCV 2021.

[2021.05.15] Our extension of RandLA-Net is accepted by TPAMI.

[2021.04.09] We are going to organize The 1st Challenge on Point Cloud Understanding at ICCV 2021.

[2021.03.01] Our papers SensatUrban and SpinNet are accepted by CVPR 2021.

[2021.02.28] Our paper RadarLoc is accepted by ICRA 2021.

[2020.11.28] We organize a tutorial about 3D point cloud learning at 3DV 2020.

[2020.02.27] Our paper RandLA-Net for 3D semantic segmentation is accepted by CVPR 2020.

[2019.09.03] Our paper 3D-BoNet is accepted as a spotlight at NeurIPS 2019.

[2019.08.16] Our paper AttSets is accepted by IJCV.

OGC: Unsupervised 3D Object Segmentation from Rigid Dynamics of Point Clouds
Z. Song, B. Yang
Advances in Neural Information Processing Systems (NeurIPS), 2022
arXiv / Video/ Code

We introduce the first unsupervised 3D object segmentation method on point clouds.

Promising or Elusive? Unsupervised Object Segmentation from Real-world Single Images
Y. Yang, B. Yang
Advances in Neural Information Processing Systems (NeurIPS), 2022
arXiv / Project Page/ Code

We systematically investigate the effectiveness of existing unsupervised models on challenging real-world images.

GRF: Learning a General Radiance Field for 3D Representation and Rendering
A. Trevithick, B. Yang
IEEE International Conference on Computer Vision (ICCV), 2021
arXiv / Code

We introduce a simple implicit neural function to represent complex 3D geometries purely from 2D images.

Learning Semantic Segmentation of Large-Scale Point Clouds with Random Sampling
Q. Hu, B. Yang*, L. Xie, S. Rosa, Y. Guo, Z. Wang, N. Trigoni, A. Markham
IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI), 2021 (IF=16.39)
arXiv/ IEEE Xplore/ Code
(* indicates corresponding author)

The journal version of our RandLA-Net. More experiments and analysis are included.

Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds
B. Yang, J. Wang, R. Clark, Q. Hu, S. Wang, A. Markham, N. Trigoni
Advances in Neural Information Processing Systems (NeurIPS), 2019 (Spotlight, 200/6743)
arXiv / ScanNet Benchmark / Reddit Discussion / News: (新智元, 图像算法, AI科技评论, 将门创投, CVer, 泡泡机器人)/ Video/ Code

We propose a simple and efficient neural architecture for accurate 3D instance segmentation on point clouds. It achieves the SOTA performance on ScanNet and S3DIS (June 2019).

Learning to Reconstruct and Segment 3D Objects

B. Yang
Oxford Research Archive (PDF), 2020; (News: 机器之心报道)

Thesis committee (Transfer/Confirmation/Viva):
Alessandro Abate, Andrew Davison, Pawan Kumar, Andrew Zisserman.

This thesis aims to understand scenes and the objects within them by learning general and robust representations using deep neural networks, trained on large-scale real-world 3D data. In particular, the thesis makes three core contributions from object-level 3D shape estimation from single or multiple views to scene-level semantic understanding.

[2022.12] Invited talk about Unsupervised 2D/3D Object Segmentation at TechBeat forum.

[2022.06] Invited talk about 3D Scene Reconstruction, Decomposition and Manipulation at Xiamen University.

[2021.10] Invited talk about 3D Representation Learning at GAMES Webinar.

[2021.04] Invited talk about Beyond Supervised Learning for 3D Representations at a CSIG workshop.

[2020.10] Invited talk about 3D Scene Understanding at Wonderland AI Summit. Check out the trailer.

[2020.09] Invited talk about 3D Point Cloud Segmentation at MFI 2020.

[2020.03] Invited talk about our RandLA-Net and 3D-BoNet at Shenlan. Here are the Video and Slides.

[2018 -] Regularly review papers for top-tier conferences and journals in machine learning and computer vision.

Fall Term, 2022:    Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Spring Term, 2022:    Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Fall Term, 2021:    Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Spring Term, 2021:    Machine Learning and Data Analytics (The Hong Kong Polytechnic University).

Hilary Term, 2019:    Knowledge Representation & Reasoning (University of Oxford).

Michaelmas Term, 2018 & 2017:    Machine Learning (University of Oxford).

Spring Term, 2014:    C++ Programming (The University of Hong Kong).

Qingyong Hu (Oct 2018 - ):    Department of Computer Science at University of Oxford.

Alexander Trevithick (Oct 2019 - Mar 2021):    Now PhD at UCSD.

Jianan Wang (May - Dec 2018):    Now with Google DeepMind.

Zihang Lai (Oct 2017 - Mar 2018):    Now PhD at CMU.

In my free time, I like playing tennis on lawns, clays, and hard surfaces. I also like to fly drones for landscape photography. Here's a video over the historic Oxford [Youtube, 腾讯视频], and another video for the scenic Lake District [Youtube]. Remember to turn up the volume for the background music.

Last update: 2023.01. Thanks.