Hao Zhao

Hao Zhao

Auto-generated profile, based on publicly available information

Hao Zhao

Institute for AI Industry Research (AIR), Tsinghua University, Beijing, China

Research Focus & Metrics

Academic Career Length

10Years (2017-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
4 February 2024
Verified
Book Chapter
Closed access
PDF file
ZWZirui Wu
Zirui Wu
TLTianyu Liu
Tianyu Liu
LLLiyi Luo
Liyi Luo
ZZZhide Zhong
Zhide Zhong
JCJianteng Chen
Jianteng Chen
HZHao Zhao
Hao Zhao

17 January 2020 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
CYChenggang Yan
Chenggang Yan
BSBiyao Shao
Biyao Shao
HZHao Zhao
Hao Zhao
RNRuixin Ning
Ruixin Ning
YZYongdong Zhang
Yongdong Zhang
FXFeng Xu
Feng Xu
3D layout is crucial for scene understanding and reconstruction, and very useful in applications like real estate and furniture design. In this paper, we propose a fully automatic solution to estimate 3D layout of an indoor scene from a single 2D image. Our technique contains two key components. Firstly, we train a neural network that directly estimates room structure lines from the input image Secondly, we propose a novel technique to automatically identify the layout topology of an input image, followed by a nonlinear optimization with equality constraints to estimate the final 3D layout of a scene. Based on our knowledge, this is the first fully automatic technique to achieve single image-based 3D layout estimation of an indoor scene. We evaluate our method on the public datasets LSUN, Hedau and 3DGP and the results show that the proposed method achieves accurate 3D layout reconstruction on various images with different layout topologies.

21 August 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Open access
ZJZhou Jiang
Zhou Jiang
ZZZhenxin Zhu
Zhenxin Zhu
PLPengfei Li
Pengfei Li
HGHuan-Ang Gao
Huan-Ang Gao
TYTianyuan Yuan
Tianyuan Yuan
HZHao Zhao
Hao Zhao
Autonomous vehicles are gradually entering city roads today, with the help of high-definition maps (HDMaps). However, the reliance on HDMaps prevents autonomous vehicles from stepping into regions without this expensive digital infrastructure. This fact drives many researchers to study online HDMap generation algorithms, but the performance of these algorithms at far regions is still unsatisfying. We present P-MapNet, in which the letter P highlights the fact that we focus on incorporating map priors to improve model performance. Specifically, we exploit priors in both standard-definition map (SDMap) and HDMap. On one hand, we extract weakly aligned SDMap from OpenStreetMap and encode it as an additional conditioning branch. Despite the misalignment challenge, our attention-based architecture adaptively attends to relevant SDMap skeletons and significantly improves performance. On the other hand, we exploit a masked autoencoder to capture the prior distribution of HDMap, which can serve as a refinement module to mitigate occlusions and artifacts. We benchmark on the nuScenes and Argoverse2 datasets.

14 January 2022 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
XCXiaoxue Chen
Xiaoxue Chen
HZHao Zhao
Hao Zhao
GZGuyue Zhou
Guyue Zhou
YZYa-Qin Zhang
Ya-Qin Zhang
3D scene understanding from point clouds plays a vital role for various robotic applications. Unfortunately, current state-of-the-art methods use separate neural networks for different tasks like object detection or room layout estimation. Such a scheme has two limitations: 1) Storing and running several networks for different tasks are expensive for typical robotic platforms. 2) The intrinsic structure of separate outputs are ignored and potentially violated. To this end, we propose the first transformer architecture that predicts 3D objects and layouts simultaneously, using point cloud inputs. Unlike existing methods that either estimate layout keypoints or edges, we directly parameterize room layout as a set of quads. As such, the proposed architecture is termed as P(oint)Q(uad)-Transformer. Along with the novel quad representation, we propose a tailored physical constraint loss function that discourages object-layout interference. The quantitative and qualitative evaluations on the public benchmark ScanNet show that the proposed PQ-Transformer succeeds to jointly parse 3D objects and layouts, running at a quasi-real-time (8.91 FPS) rate without efficiency-oriented optimization. Moreover, the new physical constraint loss can improve strong baselines, and the F1-score of the room layout is significantly promoted from 37.9% to 57.9%. 1 1Code and models can be accessed at https://github.com/OPEN-AIR-SUN/PQ-Transformer.

10 November 2022 in ISPRS Journal of Photogrammetry and Remote Sensing
Verified
Journal Article
Closed access
BTBeiwen Tian
Beiwen Tian
LLLiyi Luo
Liyi Luo
HZHao Zhao
Hao Zhao
GZGuyue Zhou
Guyue Zhou

12 November 2019 in International Journal of Computer Vision
Verified
Journal Article
Closed access
PDF file
HZHao Zhao
Hao Zhao
MLMing Lu
Ming Lu
AYAnbang Yao
Anbang Yao
YCYurong Chen
Yurong Chen
LZLi Zhang
Li Zhang