The University of Southern Maine (USM) is celebrating the success of three research papers co-authored by graduate students with faculty members Drs. Yuqi Song and Xin Zhang, all of which were recently accepted at prestigious international conferences. Two of the papers are student-led, highlighting USM’s growing culture of graduate research and innovation.
Sean Fletcher, a USM graduate student, is the first author of “Deep Learning for Taxol Exposure Analysis: A New Cell Image Dataset and Attention-Based Baseline Model,” which was accepted as a full paper at the 2025 IEEE International Conference on Future Machine Learning and Data Science (FMLDS 2025). This research introduces a publicly available microscopy image dataset of C6 glioma cells under different concentrations of taxol, a chemotherapy drug, and proposes a baseline model, ResAttention-KNN, which integrates ResNet-50, attention mechanisms, and k-NN to achieve robust classification performance.
Deiby Wu Lee, another USM graduate student, is the first author of “An Efficient CNN with Adaptive Loss for Binocular Depth Estimation,” which was accepted to the 2nd International Workshop on Edge Intelligence and Vehicular Networks (EIVN 2025). The work presents a lightweight convolutional neural network and a novel adaptive loss function designed for real-time binocular depth estimation, achieving competitive accuracy with significantly reduced computational cost.
Wu Lee also contributed as third author on “A Review of Vision-Based Depth Estimation: Current Methods and Future Directions,” also accepted at EIVN 2025. This survey categorizes deep learning-based methods for depth estimation, benchmarks current state-of-the-art approaches, and outlines future research challenges for building more efficient and reliable systems.
“These acceptances demonstrate the high level of scholarship and innovation our graduate students are achieving,” said Professors Song and Zhang. “We are proud to see their work gaining recognition on the international stage.”
Together, these accomplishments showcase USM’s commitment to advancing research in artificial intelligence, computer vision, and machine learning, while providing graduate students with the opportunity to lead and contribute to impactful projects.