Jinjie (Wayne) Mai
Jinjie (Wayne) Mai

About Me

Jinjie Mai is an M.S./Ph.D. student at KAUST, fortunately under the supervision of Prof. Bernard Ghanem. He has a broad research interest in 3D vision including topics like localization, reconstruction, and generation. He has published many papers at top-tier AI conferences such as CVPR/ICCV/ECCV/NeurIPS/ICLR. He has collectively earned over 1,000 citations. His current research now focuses on 4D reconstruction and generation.

Interests
  • 3D and 4D Reconstruction
  • 3D and 4D Generation
  • 3D and 4D Understanding
  • Diffusion Models
Education
  • KAUST logo

    KAUST

    MSc Computer Science

  • Sun Yat-Sen University logo

    Sun Yat-Sen University

    BEng Computer Science and Technology

Featured Publications
Sora3R: Can Video Diffusion Model Reconstruct 4D Geometry ?

Sora3R: Can Video Diffusion Model Reconstruct 4D Geometry ?

Jinjie Mai, Wenxuan Zhu, Haozhe Liu, Bing Li, Cheng Zheng, Jürgen Schmidhuber, Bernard Ghanem
In arXiv preprint · 2025

We repurpose video diffusion models to reconstruct 4D geometry (4D pointmaps) from monocular video.

TrackNeRF: Bundle Adjusting NeRF from Sparse and Noisy Views via Feature Tracks

TrackNeRF: Bundle Adjusting NeRF from Sparse and Noisy Views via Feature Tracks

Jinjie Mai, Wenxuan Zhu, Sara Rojas, Jesus Zarzar, Guocheng Qian, Abdullah Hamdi, Bing Li, Silvio Giancola, Bernard Ghanem
In ECCV · 2024

We introduce global bundle adjustment for NeRFs, significantly improving the quality of novel view synthesis from sparse views with noisy poses or even without poses.

Vivid-ZOO: Multi-View Video Generation with Diffusion Model

Vivid-ZOO: Multi-View Video Generation with Diffusion Model

Bing Li, Cheng Zheng, Wenxuan Zhu, Jinjie Mai, Biao Zhang, Peter Wonka, Bernard Ghanem
In NeurIPS · 2024

We propose multiview video diffusion models that generate high-quality multi-view videos for dynamic objects from text.

Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors

Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors

Guocheng Qian, Jinjie Mai, Abdullah Hamdi, Jian Ren, Aliaksandr Siarohin, Bing Li, Hsin-Ying Lee, Ivan Skorokhodov, Peter Wonka, Sergey Tulyakov, Bernard Ghanem
In ICLR · 2024

We propose to use both 2D and 3D diffusion models through SDS optimization to generate high-quality, textured 3D meshes from a single image in the wild.

EgoLoc: Revisiting 3D Object Localization from Egocentric Videos with Visual Queries

EgoLoc: Revisiting 3D Object Localization from Egocentric Videos with Visual Queries

Jinjie Mai, Abdullah Hamdi, Silvio Giancola, Chen Zhao, Bernard Ghanem
In ICCV (Oral) · 2023

We propose EgoLoc, a novel approach for localizing 3D objects from egocentric videos, achieving SOTA performance on the Ego4D VQ3D task with 87.12% success rate over previous SOTA of only 8%.

PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies

PointNeXt: Revisiting PointNet++ with Improved Training and Scaling Strategies

Guocheng Qian, Yuchen Li, Houwen Peng, Jinjie Mai, Hasan Abed Al Kader Hammoud, Mohamed Elhoseiny, Bernard Ghanem
In NeurIPS · 2022

We revisit PointNet++ and propose PointNeXt, a next-generation PointNet architecture.

Erasing Integrated Learning : A Simple yet Effective Approach for Weakly Supervised Object Localization

Erasing Integrated Learning : A Simple yet Effective Approach for Weakly Supervised Object Localization

Jinjie Mai, Meng Yang, Wenfeng Luo
In CVPR (Oral) · 2020

We propose EIL, a novel adversarial erasing technique for weakly supervised object localization.

Awards
VQ3D Challenge, 1st Winner (ECCV 2022)
2nd International Ego4D Workshop @ ECCV 2022 ∙ September 2022
  • Acted as first author and team leader
  • Achieved 1st place in the VQ3D Challenge
  • Technical report available on arXiv
VQ3D Challenge, 1st Winner (CVPR 2023)
3rd International Ego4D Workshop @ CVPR 2023 ∙ June 2023
  • Acted as first author and team leader
  • Achieved 1st place in the VQ3D Challenge
  • Technical report available on arXiv
Second Place in Men’s Doubles
SYSU CS Department Badminton Tournament ∙ May 2017
Achieved second place in the men’s doubles category.
Third Place in Men’s Doubles
SYSU CS Department Badminton Tournament ∙ May 2018
Achieved third place in the men’s doubles category.
F6 Champion in Men’s Doubles
Jeddah SBC Badminton Tournament ∙ February 2024
Champion in Men’s Doubles at the Jeddah SBC badminton tournament.
Services
Teaching Assistant :
  • Deep Learning for Visual Computing, CS 323, KAUST
Reviewer :
  • Conferences: CVPR, ICCV, ECCV, NeurIPS, ICML, ICLR
  • Journals: ACM TOMM, IEEE Transactions on Multimedia, IEEE TIP
Co-organizer :
  • 2nd 3DMV: Learning 3D with Multi-View Supervision, CVPR 2024 Workshop
(2024)
Featured Blogs
Dolomites

Dolomites

Two steps from heaven.
avatar
Jinjie (Wayne) Mai
15 mins read
Paris | 巴黎

Paris | 巴黎

Paris, it’s a way of life. | 巴黎, 作爲一种生活方式.
avatar
Jinjie (Wayne) Mai
30 mins read