About

Computer Vision engineer with extensive experience of around 2 years in designing and developing multiple large-scale products in Computer Vision and Deep Learning domain, in an end-to-end manner. As a Pattern Recognition enthusiast,I also have some experience in NLP(Natural Language Processing) and ASR(Automatic Speech Recognition). Well-versed with python, c, c++, data-structures and algorithms Experienced with frameworks including Pytorch, Tensorflow, Keras Skilled in Image Processing, Computer Vision, and Deep Learning, software development skills High familiarity with libraries such as OpenCV, Pandas, Numpy, Scikit, SkLearn, Flask, DLib, QT and version control tool such as Git. Exposed working on Unix and Microsoft OS, and exposure to Raspbian OS and embedded hardware such as Raspberry Pi and Nvidia Jetson Nano. Exposed to model optimization techniques using TensorRT, OpenVino, Tflite and other frameworks. Exposed to web services such as AWS, Google cloud, Azure and Heroku Computer Vision applications : object detection, object classification, face detection and recognition, handwriting detection, OCR, object tracking ,image search, motion analysis, key-frame extraction. NLP applications : Information retrieval and summarizer Deep Learning models : VGG-16, Resnet, InceptionNet, Mobilenet, Yolo, ssd-resnet10, ssd-mobilenetv2, U-net, Siamese Network Image Processing algorithms : HoG, SURF, SIFT, LBP, Canny Edge Detector, Optical Flow, Filters, background-subtraction Machine Learning algorithms : SVM, Decision Trees, KMeans, KNN, Dbscan, TF-IDF, LDF, BoV My research papers addresses the problem of activity recognition in cricket trimmed videos, and describes how we used semi-supervised approach and integrating ideas from NLP and image processing to tackle the problem. Having worked on multiple development and research projects, I find myself well equipped with research and project management skills that are necessary for AI based roles, and hence, I'm looking forward for such roles where I can create a positive impact and solve real world problems.