Multi-Cam-ReID
Published:
Developed a robust multi-object tracking system with custom tracking algorithms, including advanced velocity modeling for enhanced performance in occlusion scenarios. 
Projects
Spatial-AI
Published:
Enhanced depth estimation and generated more accurate disparity maps utilizing the OAK-D Pro camera, employing a methodology that combines traditional knowledge of stereo vision with state-of-the-art deep learning techniques. 
Vision-Beyond-Limits
Published:
Extracted and classified buildings based on damage levels from the post-earthquake imagery training set, assigning color codes to each category. Implemented multi-label classification model for accurate classification in our network training process. 
Image Grouping
Published:
Developed an image grouping algorithm that effectively segregated similar images based on extracted features through the implementation of clustering techniques.