Research Foundation

I am interested in exploring foundational methods for learning features and representations, enabling applications across deep learning, computer vision, and multimedia. 

Selected Publications: 

• TIACam: Auto-augmentation learns non-semantic variations, and text anchors semantic meaning, enabling invariant feature learning. [https://arxiv.org/abs/2602.18863] 

• Text-guided invariant feature learning that leverages multimodal alignment.  [https://arxiv.org/abs/2503.13805]

• Padding Module: Learning the Padding in Convolutional Neural Networks. [https://arxiv.org/abs/2301.0460]

• Perspective Transformation Layer.  [https://arxiv.org/abs/2201.05706]

• Deep Morphological Neural Networks.  [https://arxiv.org/abs/1909.01532]