Qi Ma - Homepage

Qi Ma (马麒)

I am a second-year PhD Student at Computer Vison Lab in ETH Zurich, Switzerland under the supervision of Prof. Dr. Ender Konukoglu, and joint supervised by Prof. Dr. Luc Van Gool and Dr. Danda Pani Paudel in INSAIT, Soifa University .

Before I received my Master Degree from ETH Zurich and receive the honor of ETH Medal: Outstanding master thesis under the supervision of Prof. Dr. Luc Van Gool. After the Master, I worked as a Research Assistant at INSAIT for 3 months.

News

02/2025 Our paper CityLoc: 6DoF Pose Distributional Localization for Text Descriptions in Large-Scale Scenes with Gaussian Representation is released on Arxiv.
01/2025 Our paper ShapeSplat: A Large-scale Dataset of Gaussian Splats and Their Self-Supervised Pretraining was accepted at 3DV 2025 Oral.
11/2024 Our paper Video Foundation Model for Medical 3D Segmentation was accepted at MICCAI 2024 ToothFairy Workshop.
09/2024 Our paper Implicit Zoo: A Large-Scale Dataset of Neural Implicit Functions for 2D Images and 3D Scenes was accepted at Neurips 2024 , see you in Vancouver!.
08/2024 Our paper ShapeSplat: A Large-scale Dataset of Gaussian Splats and Their Self-Supervised Pretraining is released on Arxiv.
08/2024 Score: 87.3/100 in MICCAI Workshop of Multi-Structure Segmentation in CBCT Volumes: Toothfairy2 Challenge!
05/2023 Our paper Implicit Zoo submit to Neurips Dataset and Benchmark Track
04/2024 Career Update: I start working as a PhD Student at ETH BMIC!
02/2024: The extension of my master thesis Continuous Pose for Monocular Cameras in Neural Implicit Representation got accepted to CVPR 2024
12/2023 Career Update: I start working as a Research Assistant at INSAIT, Sofia!
03/2023 Our paper Deformable Neural Radiance Fields using RGB and Event Cameras is accepted to ICCV 2023.

Research

CityLoc: 6DoF Pose Distributional Localization for Text Descriptions in Large-Scale Scenes with Gaussian Representation
Qi Ma, Runyi Yang, Bin Ren^✉ , Nicu Sebe, Ender Konukoglu, Luc Van Gool Danda Paudel, paper | code |

Text based Pose Distributional Localization for Large-Scale Scenes with Gaussian Representation.

ShapeSplat: A Large-scale Dataset of Gaussian Splats and Their Self-Supervised Pretraining
Qi Ma^*, Yue Li^*, Bin Ren^✉ , Nicu Sebe, Ender Konukoglu, Theo Gevers, Luc Van Gool, Danda Paudel,
Conference on International Conference on 3D Vision (3DV), 2025

(* equal contribution)
paper | project page | code |

Shapesplat: A large scale Gaussian Splatting dataset for central objects, and proposed self-supervised training Gaussian-MAE.

Implicit-Zoo: A Large-Scale Dataset of Neural Implicit Functions for 2D Images and 3D Scenes
Qi Ma, Ajad Chhatkuli, Danda Paudel, Luc Van Gool
Conference on Neural Information Processing Systems (Neurips), 2024
paper | code |

A large scale implicit function dataset for various 2D and 3D tasks

Continuous Pose for Monocular Cameras in Neural Implicit Representation
Qi Ma, Ajad Chhatkuli, Danda Paudel, Luc Van Gool
Conference on Computer Vision and Pattern Recognition (CVPR), 2024
paper | code

Using continuous prior for better pose trakcing or refinement in NeRF/NeRF-SLAM.

Deformable Neural Radiance Fields using RGB and Event Cameras
Qi Ma, Ajad Chhatkuli, Danda Paudel, Luc Van Gool
International Conference on Computer Vision (ICCV), 2023
paper | code |

4D NeRF construction using RGB and event camera.

Teaching

Teaching Assistant, Probabilistic Artificial Intelligence, Spring 2022

Academic Services

Conference Reviewer: 3DV
Journal Reviewer: IEEE TMI

template adapted from this awesome website