Dr Wanli Ma
Dr Wanli Ma is a postdoctoral research fellow at the University of Cambridge, specialising in the intersection of Artificial Intelligence and Satellite Remote Sensing (AI for Earth Observation). He received his PhD from Cardiff University (2024), where he developed label-efficient Computer Vision models within the Visual Computing group. Supervisor: Dr Oktay Karakus (remote sensing) and Prof Paul Rosin (computer vision, Fellow of the IAPR).
Dr Ma’s research tackles the critical bottleneck of ground truth data scarcity in Earth Observation by developing label-efficient computer vision frameworks such as semi-supervised learning, active learning, and multimodal fusion. He bridges the gap between theoretical AI and its deployment in complex planetary environments. His work translates directly into high-stakes applications, including rapid disaster assessment for floods and hurricanes, marine monitoring, and autonomous urban mapping.
Beyond his research, he actively contributes to the academic community. He is a reviewer of top-tier artificial intelligence and remote sensing journals, including IEEE Transactions on Image Processing (T-IP), and IEEE Transactions on Neural Networks and Learning Systems (T-NNLS) , International Journal of Applied Earth Observation and Geoinformation (JAG) and IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (J-STARS). Alongside his research, he has over two years of teaching experience in computer science discipline, including postgraduate module “Data Science”, undergraduate module “Object Orientation, Algorithms and Data Structures” and summer courses “Python Programming” and “Practical AI”.
Select publications
- Diversenet: Decision diversified semi-supervised semantic segmentation networks for remote sensing imagery. W. Ma, O. Karakuş and P. L. Rosin, IEEE J. Sel. Top. Appl. Earth Obs. Remote Sens. 18, 14258. (2025)
- Patch-GAN Transfer Learning with Reconstructive Models for Cloud Removal, W. Ma, O. Karakuş and P. L. Rosin, IEEE Int. Geosci. Remote Sens. Symp., 6275 (2025)
- Knowledge Distillation for Road Detection Based on Cross-Model Semi-Supervised Learning, W. Ma, O. Karakuş and P. L. Rosin, IEEE Int. Geosci. Remote Sens. Symp., 8173 (2024)
- Confidence Guided Semi-Supervised Learning in Land Cover Classification, W. Ma, O. Karakuş, and P. L. Rosin, IEEE Int. Geosci. Remote Sens. Symp., 5487 (2023)
- High-precision density mapping of marine debris and floating plastics via satellite imagery, Booth, H., Ma, W. and Karakuş, O., Sci Rep 13, 6822 (2023).
- AMM-FuseNet: Attention-based multi-modal image fusion network for land cover mapping, W. Ma, O. Karakuş and P. L. Rosin, Remote Sens 14, 4458. (2022)
- Exploiting the Dual-Tree Complex Wavelet Transform for Ship Wake Detection in SAR Imagery, W. Ma, A. Achim, and O. Karakuş, IEEE Int. Conf. Acoust. Speech Signal Process, 1530-1534 (2021)
Select awards
- Ranked No. 3 MVEO Competition in BMVC 2025
- Ranked No. 2 MVEO Competition in BMVC 2023
- IEEE Geoscience and Remote Sensing Society Travel Grant 2025
- IEEE Geoscience and Remote Sensing Society Travel Grant 2024
- 2nd Prize in National College Students Photoelectric Design Competition 2018
- 1st Prize in China International Aircraft Design (CADC) Competition 2016
- Master’s degree with Distinction 2020
- Honours Bachelor’s Graduate 2019
Further links
Homepage: https://wanlima-cardiff.github.io
