Shivin Diwakar

Junior Research Engineer @Staqu Technologies

Gurugram, IN

Computer Vision Tasks: Responsible for developing and deploying various detection & classification tasks/projects pipelines using yolov5, v6, v7, v8, yolox, yolov9, yolov10, yolov11 (for detection), ViTs (vision transformers), DeITs (Data-efficient Image Transformers)(for classification) & docker containers (deployment). •Detection Tasks: Developed & deployed projects for detecting various entities like person, bag, box, face, table, animal, crockery, human-fall using yolov5, yolov9, yolov10.•Forensics Tasks: Developed projects pipelines for image forgery Detection, deepfake detection in video, image & audio.•Optical Flow: Developed tasks involving motion detection like, escalator motion & wrong-lane driving.•Action Recognition (video-moment-retrieval): Developed projects for recognising & retrieving various actions like cleaning, fighting, waving-hand, smoke, frisking, eating, drinking.•Classification Tasks: Developed & deployed classifiers for various use-cases like, classifying whether the person is wearing hat or not, similar tasks include: brand(company) classifiers, vehicle classifiers, uniform/jacket classifier, face & gender classifiers, using Vision Transformers, Data-efficient Image Transformers & ResNet.•Image Processing Tasks: Developed projects for detecting & measuring distances between various entities like distance between cars and social distancing, Optical Character Recognition, and various image processing tasks,•Miscellaneous Tasks: Researching & developing projects involving LLMs & Generative AI for various modalities.•Natural Language Processing: Worked on tasks like, Text-to-Image Retrieval (CLIP, BLIP, ALIP models), Sketch-based Retrieval

Guru Gobind Singh Indraprastha University

Bachelor of Technology - BTech

St. Cecilia's Public School

High School

Delhi University

Bachelor of Engineering - BE

Netaji Subhas Institute of Technology

Bachelor of Engineering - BE