OpenCV Jobs

I need a computer-vision pipeline that can ingest an uploaded logo image (PNG or JPG) together with a video file and then return every instance of that logo throughout the clip. The videos that come in will be MP4, AVI, or MOV, and the solution must run smoothly. Here’s how I picture the workflow: • The script accepts the logo and video paths from a simple CLI or minimal API endpoint. • It scans each frame, flags every logo hit, and outputs a tidy JSON or CSV that lists, for each detection, the start timestamp, on-screen duration, and bounding-box coordinates. • No extra training or fine-tuning whenever a new logo is supplied—the detector should work right away with whatever image is uploaded. I’m after well-commented, maintainable code, and I&rsquo...

Project Title: Computer Vision Expert for Eye Motion Detection (Mobile App - Closed-Eyelid Tracking) Project Overview We are looking for a highly skilled Computer Vision / AI Developer to build a specialized motion detection module for a mobile application. The core objective is to detect and quantify eyeball movements by analyzing video footage of a subject’s closed eyelids. Technical Challenge: "Bulge Tracking" Unlike standard pupil tracking, this project requires analyzing the movement of the eyeball "bulge" underneath the closed eyelid. The solution must be hardware-agnostic, working with standard smartphone RGB cameras under normal lighting conditions, without the need for specialized IR sensors. Key Deliverables & Requirements 1. Motion Detection: Identi...

# Техническое задание ## Разработка парсера урона для Lineage II (High Five) ### 1. Общая информация Требуется разработать десктопное приложение для **Windows**, которое в реальном времени анализирует урон в игре **Lineage II (High Five)**. Программа должна работать поверх окна игры и автоматически считывать числовые значения урона, которые появляются над персонажами во время боя. Основная цель — получать статистику нанесённого и полученного урона во время PvP (например, на олимпиаде). --- ### 2. Основной принцип работы Программа должна: 1. Определять координаты окна игры **Lineage II** (клиент работает в оконном режиме). 2. Захватывать заданную область экрана, где появляются цифры урона. 3. Выполнять обработку изображения (фильтрация цветов / выделение цифр). 4. Распознават...

Expert Mobile Developer: Infrared Eye-Tracking & Nystagmus Detection (Computer Vision). Project Overview: We are seeking a high-level Mobile Developer/Computer Vision Engineer to build a specialized module for a vertigo diagnostic support application for smartphones. The core task is to capture and analyze Nystagmus using the device's internal infrared sensors. Scientific Context: Nystagmus is an involuntary, rhythmic oscillation of the eyes caused by vestibular (inner ear) dysfunction. While vertigo is the subjective sensation of spinning, nystagmus is the objective clinical evidence. Capturing this in total darkness is essential to prevent "visual fixation" which can suppress the symptoms. Key Responsibilities: 1. Mobile IR Capture: Access internal IR sensors (e.g., i...

I want to replace paper timesheets with an automatic clock-in / clock-out system that identifies staff by face the moment they pass our main entrance camera. As soon as a recognised face crosses the frame, the system should log the arrival or departure time against that employee’s profile and store the record in an attendance database I can export to CSV or Excel. Architecture is open: I’m comfortable with a cloud, on-prem, or hybrid deployment—whichever you feel offers the best balance of accuracy, speed and privacy given a single entrance camera. No links to my HR or payroll software are needed for now, so a self-contained dashboard or lightweight web portal that lets me view, edit and download daily, weekly and monthly reports is perfect. Key deliverables &bull...

I need a complete Automatic Number Plate Recognition solution that accurately reads both car and motorcycle plates during daylight hours. Once a plate is captured, the system must do two things automatically: log the full plate data (image, text, timestamp) into a database and trigger an immediate alert through a simple webhook or REST call. My ideal stack is Python with OpenCV or a comparable vision framework, paired with a modern detection model such as YOLOv8-seg, but I am open to alternatives if you can demonstrate equal or better accuracy. You are free to use an off-the-shelf OCR engine (Tesseract, EasyOCR, PaddleOCR) or a custom-trained model, as long as you meet the accuracy target. Key deliverables • Optimized plate-detection and OCR pipeline tuned for daylight footage &bu...

Computer Vision Developer – Sports Video Analysis (Mobile-Based) Description: We are looking for a skilled computer vision developer to assist with building a prototype system that analyses sports video captured from smartphones. The goal is to detect and track fast-moving objects, identify key events within video footage, and generate structured outputs (e.g. timestamps, event detection, basic classifications). This is an early-stage prototype, so we are looking for someone who can work quickly, think practically, and help shape the technical approach. Key Responsibilities: Develop object detection and tracking models for fast-moving objects (e.g. ball tracking) Process video input from 1–2 mobile camera angles Detect key events (e.g. impacts, direction changes, motion pa...

I’m assembling a small, highly-skilled development squad to finish a platform that sees the world, talks to external services that never published their docs, and then deploys itself at scale. Python sits at the heart of the codebase, so deep confidence with modern Python, virtual environments and dependency management is essential. Most of the heavy lifting in computer vision happens through OpenCV and complementary deep-learning libraries; the pipeline must handle object detection, image processing and facial recognition with equal reliability and speed. If you also think in C++ or Java when performance demands it, that flexibility will be welcome. Beyond vision, we have to collect data from third-party applications that do not provide friendly endpoints. That means reverse-e...

I have a proprietary image dataset and I need a seasoned computer-vision engineer to turn it into an end-to-end solution that can 1) recognise each image, 2) detect all relevant objects inside it, and 3) optionally segment those objects when that adds value to the overall result. The primary goal is accurate detection and classification, but I’d like the system to be flexible enough to switch on pixel-level segmentation whenever it boosts performance. Here is what I’m after: • A clean Python pipeline (PyTorch or TensorFlow preferred, with OpenCV for preprocessing) that ingests my custom images, trains suitable models, and exposes an easy inference script. • Clear evaluation: confusion matrices, mAP/IoU scores, and plots of loss/accuracy over epochs so I can judge...

I need a lightweight, computer-vision bot that relies on on-screen image recognition rather than Selenium or similar browser-automation libraries. The workflow is straightforward: the bot launches a local browser window, navigates to a specific site I will provide, and—through visual cues—automatically clicks the required elements, fills in form fields, and finally extracts the resulting text, images, and tabular data that the page returns. The process must loop continuously so it can watch for page changes and react in real time. A small delay between cycles is fine as long as the interaction remains smooth and does not overload the system or the target site. Robust error handling is important; if a button is missing or the page layout shifts, the bot should retry gracefull...

This project is a full stack system built around a vision based browser automation bot. The bot visits three target websites, fills multi-step forms on each of them using data provided by an operator, scrapes the results returned after each submission, and stores everything in a database. The operator interacts with the system through a web interface and never touches the bot or the code directly. The system has five parts that all need to be built and connected. A React frontend is where the operator submits input data and monitors what the bot is doing in real time. A FastAPI backend sits behind the frontend and handles communication between the operator, the bot, and the database. Redis acts as the message layer between the backend and the bot, passing events like run triggers, status u...