Challenges


Subject: Invitation to MMT2025 LiDAR Point Cloud Challenges

Dear Distinguished Researcher,

The 13th International Conference on Mobile Mapping Technology (MMT2025) will be held from June 20-22, 2025, in Xiamen, China. This prestigious event is organized by the International Society of Photogrammetry and Remote Sensing (ISPRS) and hosted by Xiamen University's Spatial Sensing and Computing Laboratory in collaboration with ISPRS Working Groups I/2, I/8, and II/2.

In conjunction with MMT2025, we are organizing two data challenges focused on LiDAR point clouds-based transportation scene understanding. These challenges aim to advance cutting-edge technologies in this rapidly evolving field.

Challenge 1: Point Cloud Scene Understanding
This challenge includes three distinct tracks:

  1. Railway Scene Point Cloud Semantic Segmentation: Participants will develop semantic segmentation models based on the WHU-Railway3D dataset, which covers urban, rural, and plateau railway environments with 4.6 billion points across 11 semantic categories.
  2. Road Scene Point Cloud Semantic/Instance Segmentation: Using the WHU-Urban3D dataset, participants will develop algorithms for complex urban point cloud semantic and instance segmentation, with over 600 million points covering 19 semantic categories.
  3. Road Scene Lane Mapping: This track challenges participants to extract lane lines and their semantic attributes directly from Mobile Laser Scanning point clouds using the WHU-Lane dataset, which includes 98 kilometers of high-resolution road infrastructure data.

Further details: https://traffic3dchallenge.github.io/

Contact Email: traffic3dchallenge@outlook.com

Challenge 2: 3D Scene Object Perception
This challenge consists of two tracks:

  1. Cross-Mechanism Domain Adaptation 3D Object Detection: Participants must train detectors on point clouds from 128-beam or 32-beam mechanical LiDARs and, without any target-domain labels, generalise them to a hidden solid-state LiDAR test set for cross-mechanism domain adaptation 3D object detection.
  2. LiDAR-4D Radar Fusion for Cooperative 3D Object Detection: Participants are required to train a 3D object detector using cooperative perception data from multimodal sources, including LiDAR point clouds and 4D radar point clouds.

Further details: https://object3dd.github.io/

Contact Email: xiaqiming@stu.xmu.edu.cn

Important Dates:

  • Submission Deadline (both challenges): May 30, 2025, 24:00 Beijing Time
  • Results Notification: June 5, 2025

Should you have any questions or require further information, please do not hesitate to contact us.

Kind regards,
MMT2025 Organizing Committee
Conference website: https://mmt2025.xmu.edu.cn/2025/

isprs

  Important Dates

All deadlines are 23:59 Pacific Time (PT). No extensions will be granted.

Deadline for Workshop proposal March 1, 2025
April 1, 2025
Deadline for Full paper March 1, 2025
April 1, 2025
Deadline for Abstracts February 22, 2025
April 1, 2025
Notification of Acceptance March 22, 2025
May 1, 2025
Deadline for Camera Ready April 20, 2025
May 20, 2025
Deadline for Early Registration April 1, 2025
May 12, 2025

  Sponsors

Want to sponsor the conference? Check out Sponsor and contact mmt2025sponsor@outlook.com

  Latest News