Many interesting applications in computer vision such as homography estimation, plane detection, 3D reconstructions and motion estimation demand the ability to fit geometric models onto noisy data. This is a non-trivial task in that the scene typically consists of multiple geometric structures. Moreover, the observed data is likely to be contaminated with noise from different sources including measurement sensor noise and outliers. Our group has developed several novel algorithms to solve the robust model-fitting problem and few examples are listed below:

• Tennakoon, R., Suter, D., Zhang, E., Chin, T.J. and Bab-Hadiashar, A. Consensus Maximisation Using Influences of Monotone Boolean Functions. IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021).

• Muthu, S., Tennakoon, R., Rathnayake, T., Hoseinnezhad, R., Suter, D. and Bab-Hadiashar, A. Motion segmentation of RGB-D sequences: Combining semantic and motion information using statistical inference. IEEE Transactions on Image Processing (TIP).

• Tennakoon, R., Sadri, A., Hoseinnezhad, R. and Bab-Hadiashar, A., 2018. Effective sampling: Fast segmentation using robust geometric model fitting. IEEE Transactions on Image Processing (TIP).

• Tennakoon, R.B., Bab-Hadiashar, A., Cao, Z., Hoseinnezhad, R. and Suter, D. Robust model fitting using higher than minimal subset sampling. IEEE transactions on pattern analysis and machine intelligence (PAMI).

Medical imaging analysis can provide direct, sensitive and consistent measures for the diagnosis and understanding of diseases. Our recent research in this regard is focused on 3D CT, PET and MRI image analysis. Examples of past work include:

• Korevaar, S., Tennakoon, R., Page, M., Brotchie, P., Thangarajah, J., Florescu, C., Sutherland, T., Kam, N.M. and Bab-Hadiashar, A. Incidental detection of prostate cancer with computed tomography scans. Nature - Scientific Reports 2021.

• Tennakoon, R., Bortsova, G., Ørting, S., Gostar, A.K., Wille, M.M., Saghir, Z., Hoseinnezhad, R., de Bruijne, M. and Bab-Hadiashar, A. Classification of volumetric images using multi-instance learning and extreme value theorem. IEEE transactions on medical imaging (TMI).

• Bogunović, H., Venhuizen, F., Klimscha, S., Apostolopoulos, S., Bab-Hadiashar, A., Bagci, U., Beg, M.F., Bekalo, L., Chen, Q., Ciller, C. and Gopinath, K. RETOUCH: The retinal OCT fluid detection and segmentation benchmark and challenge. IEEE transactions on medical imaging (TMI)

• Tennakoon, R.B., Bab-Hadiashar, A., Cao, Z. and de Bruijne, M. Nonrigid registration of volumetric images using ranked order statistics. IEEE transactions on medical imaging (TMI).

3D scene construction and analysis are vital tasks in computer vision. Our group is working on several aspects of 3D vision including: Learning based stereo matching techniques and point cloud analysis. Examples include:

• Chuah, W., Tennakoon, R., Hoseinnezhad, R. and Bab-Hadiashar, A. Deep Learning-Based Incorporation of Planar Constraints for Robust Stereo Depth Estimation in Autonomous Vehicle Applications. IEEE Transactions on Intelligent Transportation Systems.

• Mukhaimar, A., Tennakoon, R., Lai, C.Y., Hoseinnezhad, R. and Bab-Hadiashar, A. Pl-net3d: Robust 3d object class recognition using geometric models. IEEE Access.

• Mukhaimar, A., Tennakoon, R., Lai, C.Y., Hoseinnezhad, R. and Bab-Hadiashar, A. Comparative analysis of 3D shape recognition in the presence of data inaccuracies. In 2019 IEEE International Conference on Image Processing (ICIP).