Vinay Sannaiah

I am currently working on my Master thesis named "Camera based instance flow for dynamic objects". The aim of the project is to verify the possibility of extending the Mask-RCNN architecture to have a flow head. Mask RCNN, which is an extension of Faster RCNN, is currently the state of the art for instance segmentation. I believe that the architecture can be extended to include an optical flow head similar to the mask head.

Expertise: 3D rendering, data preprocessing, Intel RealSense D435 sensor, GitHub, Python, C++, Pygame etc. Project: Head-pose invariant 2D/3D facial expression recognition with deep learning • Rendering of 3D textured meshes from the given object and material files for the 3D images. • Python scripts to automatically capture the 3D renderings with different yaw, pitch and angle were written. • Intel RealSense D435 camera was used to capture both RGB and depth images to prepare own dataset.

Trained on Java, SQL, Sonic ESB and Progress 4GL technologies. Developing Web services and Java message services for the interaction between the data base and the frontend. Deploying the programs and the API’s to higher level environments.

Computer Vision and Image processing

Electronics and Communication Engineering Digital Image processing Signal Processing Embedded Systems VLSI etc.