Michigan State University
Indian Institute of Technology Jodhpur
Delhi Public School Ghaziabad Vasundhara
Collected a large scale Furniture dataset with images as well as metadata. Investigated how Siamese networks can be used efficiently for assessing the style compatibility between images of furniture items. Also used a joint image-text embedding method that allows for the querying of stylistically compatible furniture items, along with additional attribute constraints based on text.
Developed a 3-D Convolution Deep Neural Network for person identification using Gait under multiple view.The proposed system is much more efficient in terms of time and space and performing better for almost all angles. Also worked on the quality estimation of finger knuckle images based on 6 different quality parameters using convolutional neural networks.