Philip Chikontwe

Philip Chikontwe is a PhD Candidate in the Medical Imaging and Signal Processing Laboratory (MISPL) under the Robotics Engineering department at DGIST.


PhD. Robotics Engineering. Daegu Gyeongbuk Institute of Science and Technology (2019-2023)

Masters Degree. Computer Science and Engineering. Chonbuk National University (2018)

Diploma. Korean Language and Literature. Sun Moon University (2016)

Bachelors Degree. Computer Science. Université Mentouri de Constantine (2015)

Diploma. French Language and Literature. Université Aboubekr Belkaid (2012)

Research Interests

Weakly supervised learning for medical/natural images (e.g. Multiple instance learning)

Learning from limited data using meta-learning (few-shot classification and segmentation)


Selected publications

Kim, Soopil, Sion An, Philip Chikontwe, and Sang Hyun Park. 2021. “Bidirectional RNN-Based Few Shot Learning for 3D Medical Image Segmentation”. Proceedings of the AAAI Conference on Artificial Intelligence 35 (3):1808-16.

Chikontwe, P., Gao, Y. & Lee, H.J. Transformation guided representation GAN for pose invariant face recognition. Multidim Syst Sign Process 32, 633–649 (2021).

E. Jung, P. Chikontwe, X. Zong, W. Lin, D. Shen and S. H. Park, "Enhancement of Perivascular Spaces Using Densely Connected Deep Convolutional Neural Network," in IEEE Access, vol. 7, pp. 18382-18391, 2019, doi: 10.1109/ACCESS.2019.2896911.

I. Ullah, P. Chikontwe and S. H. Park, "Guidewire Tip Tracking using U-Net with Shape and Motion Constraints," 2019 International Conference on Artificial Intelligence in Information and Communication (ICAIIC), 2019, pp. 215-217, doi: 10.1109/ICAIIC.2019.8669088.

I. Ullah, P. Chikontwe and S. H. Park, "Real-Time Tracking of Guidewire Robot Tips Using Deep Convolutional Neural Networks on Successive Localized Frames," in IEEE Access, vol. 7, pp. 159743-159753, 2019, doi: 10.1109/ACCESS.2019.2950263.