I am a Research Engineer at Apple, researching on-device Visual-Language & Computer Vision capabilities at Apple scale. I design visual grounding and vision-language models (VLMs) to meet latency and memory targets, and have contributed to personalized image generation for Apple Intelligence, pet detection in Photos (iOS 17), and gesture-triggered visual effects.
Prior to Apple, I was a Senior Research Scientist at Clarifai improving face detection models, a Senior Research Engineer at Vidado.ai developing large-scale image classification and document retrieval systems, and a Research Scientist at Sony US Research Center optimizing semantic segmentation.
I received my Ph.D. in EECS from Northwestern University in 2016, where I worked under the supervision of Aggelos K. Katsaggelos in the Image and Video Laboratory (IVPL) on sparse representation and deep learning for image and video reconstruction.
Focusing on image and video understanding for on-device human experiences. Working on computer vision and deep learning solutions for Apple's ecosystem.
Worked on building a comprehensive computer vision platform, developing machine learning models and algorithms for various visual recognition tasks.
Specialized in document image analysis, developing AI solutions for automated document processing and understanding.
Worked on semantic segmentation for limited resource computing devices, optimizing deep learning models for edge deployment.
Elsevier Digital Signal Processing, 2020
We propose a novel encoder-decoder neural network model called DeepBinaryMask for video compressive sensing. The proposed framework is an end-to-end model where the sensing matrix is trained along with the video reconstruction.
IEEE Signal Processing Magazine (SPM), 2018
We review the popular neural network architectures used for imaging tasks, offering some insight as to how these deep-learning tools can solve the inverse problem.
Elsevier Digital Signal Processing, 2018
A deep learning framework for video compressive sensing.
IEEE Transactions on Image Processing (TIP), 2017
A fast iterative method to address the face identification problem with block occlusions.
International Conference Image Processing (ICIP), 2013
The approach takes advantage of Multiple Measurement Vectors (MMV), seeking for significantly sparser solutions, assuming that the solution vectors have similar sparsity structure.
