Leheng Li ζδΉζ
Bio: I am a third-year Ph.D. student at The Hong Kong University of Science and Technology (Guangzhou), supervised by Prof. Ying-Cong Chen. My research interests are computer vision and autonomous driving. I received B.Sc in Mathematics from Dalian University of Technology in 2022. Previously, I was fortunate to have interned at MEGVII Technology and NIO.
Publications
OmniBooth: Learning Latent Control for Image Synthesis with Multi-modal Instruction
Leheng Li, Weichao Qiu, Xu Yan, Jing He, Kaiqiang Zhou, Yingjie Cai, Qing Lian, Bingbing Liu, Ying-Cong Chen
arxiv preprint, 2024
An image generation framework that enables spatial control with instance-level multi-modal control. Core idea: Extend ControlNet input from RGB space to latent space.
Paper / Code / Model / Project Page /
Leheng Li, Weichao Qiu, Xu Yan, Jing He, Kaiqiang Zhou, Yingjie Cai, Qing Lian, Bingbing Liu, Ying-Cong Chen
arxiv preprint, 2024
An image generation framework that enables spatial control with instance-level multi-modal control. Core idea: Extend ControlNet input from RGB space to latent space.
Paper / Code / Model / Project Page /
@article{he2024neural,
title={Neural Radiance Field in Autonomous Driving: A Survey},
author={He, Lei and Li, Leheng and Sun, Wenchao and Han, Zeyu and Liu, Yichen and Zheng, Sifa and Wang, Jianqiang and Li, Keqiang},
journal={arXiv preprint arXiv:2404.13816},
year={2024}
}
SyntheOcc: Synthesize Geometric Controlled Street View Images through 3D Semantic MPIs
Leheng Li, Weichao Qiu, Yingjie Cai, Xu Yan, Qing Lian, Bingbing Liu, Ying-Cong Chen
arxiv preprint, 2024
A diffusion model that generate images by voxel guidance. Convey data prior from SD to driving scenarios. Building blocks of Generative Simulation Model.
Paper / Code / Project Page /
Leheng Li, Weichao Qiu, Yingjie Cai, Xu Yan, Qing Lian, Bingbing Liu, Ying-Cong Chen
arxiv preprint, 2024
A diffusion model that generate images by voxel guidance. Convey data prior from SD to driving scenarios. Building blocks of Generative Simulation Model.
Paper / Code / Project Page /
@article{he2024neural,
title={Neural Radiance Field in Autonomous Driving: A Survey},
author={He, Lei and Li, Leheng and Sun, Wenchao and Han, Zeyu and Liu, Yichen and Zheng, Sifa and Wang, Jianqiang and Li, Keqiang},
journal={arXiv preprint arXiv:2404.13816},
year={2024}
}
Neural Radiance Field in Autonomous Driving: A Survey
Lei He, Leheng Li, Wenchao Sun, Zeyu Han, Yichen Liu, Sifa Zheng, Jianqiang Wang, Keqiang Li
arxiv preprint, 2024
We systematically explored the applications of NeRF within the realm of autonomous driving, encompassing perception, reconstruction, simulation, and SLAM.
Paper /
Lei He, Leheng Li, Wenchao Sun, Zeyu Han, Yichen Liu, Sifa Zheng, Jianqiang Wang, Keqiang Li
arxiv preprint, 2024
We systematically explored the applications of NeRF within the realm of autonomous driving, encompassing perception, reconstruction, simulation, and SLAM.
Paper /
@article{he2024neural,
title={Neural Radiance Field in Autonomous Driving: A Survey},
author={He, Lei and Li, Leheng and Sun, Wenchao and Han, Zeyu and Liu, Yichen and Zheng, Sifa and Wang, Jianqiang and Li, Keqiang},
journal={arXiv preprint arXiv:2404.13816},
year={2024}
}
Adv3D: Generating 3D Adversarial Examples in Driving Scenarios with NeRF
Leheng Li, Qing Lian, Ying-Cong Chen
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024 oral
We present the first exploration of modeling adversarial examples as NeRFs. Our examples demonstrate satisfactory transferability and physical realizability in driving scenarios.
Paper / Code / Project Page /
Leheng Li, Qing Lian, Ying-Cong Chen
IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2024 oral
We present the first exploration of modeling adversarial examples as NeRFs. Our examples demonstrate satisfactory transferability and physical realizability in driving scenarios.
Paper / Code / Project Page /
@article{li2023adv3d,
title={Adv3D: Generating 3D Adversarial Examples in Driving Scenarios with NeRF},
author={Li, Leheng and Lian, Qing and Chen, Ying-Cong},
journal={arXiv preprint arXiv:2309.01351},
year={2023}
}
Lift3D: Synthesize 3D Training Data
by Lifting 2D GAN to 3D Generative Radiance Field
Leheng Li, Qing Lian, Luozhou Wang, Ningning Ma, Ying-Cong Chen
Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2023
The first work to use NeRF-generated datasets to benefit downstream tasks. The datasets enjoy both photorealistic synthesis and 3D-control property, improving 3D detection performance in a Real2Sim2Real manner without domain adaptation.
Paper / Code / Project Page /
Leheng Li, Qing Lian, Luozhou Wang, Ningning Ma, Ying-Cong Chen
Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR), 2023
The first work to use NeRF-generated datasets to benefit downstream tasks. The datasets enjoy both photorealistic synthesis and 3D-control property, improving 3D detection performance in a Real2Sim2Real manner without domain adaptation.
Paper / Code / Project Page /
@InProceedings{lift3D2023CVPR,
author = {Leheng Li and Qing Lian and Luozhou Wang and Ningning Ma and Ying-Cong Chen},
title = {Lift3D: Synthesize 3D Training Data by Lifting 2D GAN to 3D Generative Radiance Field},
booktitle = {Proc. IEEE Conf. on Computer Vision and Pattern Recognition (CVPR)},
year = {2023},
}Homepage Template
I really appreciate Michael Niemeyer for providing the source code of the website template.