Data Labeler / Video Annotation Specialist for Computer Vision CCTV Project

We are looking for a highly detail-oriented Data Labeler / Video Annotation Specialist to support a computer vision and machine learning project focused on analyzing CCTV footage from real-world operational environments such as restaurants, retail locations, and similar business settings. This is not a simple “draw boxes around objects” task. We are building high-quality training data for machine learning models that need to understand people, objects, activity, movement, zones, workflow events, congestion, idle time, and operational behavior from video. The quality of your annotations will directly affect model performance, false detections, business insights, and future automation. We are looking for someone who is consistent, careful, reliable, and comfortable working with detailed labeling rules. What You’ll Be Doing You will review image and video data and apply precise annotations according to detailed labeling instructions. Typical annotation work may include: Labeling people, objects, work areas, counters, queues, equipment zones, and operational areas Drawing bounding boxes, polygons, segmentation masks, or other annotation shapes depending on the task Tracking objects or people across video frames Labeling activities and events such as movement, pickup, preparation, waiting, idle time, congestion, and workflow activity Applying consistent labels across large volumes of footage Flagging unclear, corrupted, low-quality, or unusable footage Following detailed taxonomy and labeling guidelines Reviewing your own work before submission Identifying ambiguous cases and escalating questions clearly Participating in feedback and calibration cycles to improve accuracy The Ideal Candidate You are a strong fit for this role if you: Have excellent visual attention to detail Can follow detailed instructions precisely Are comfortable working with repetitive visual data Care about accuracy and consistency Can meet agreed-upon throughput expectations without sacrificing quality Are comfortable receiving feedback and improving quickly Communicate clearly when something is ambiguous Take ownership of your work and do not submit rushed or careless annotations Prior experience with video annotation or image labeling is strongly preferred, but the most important traits are accuracy, consistency, patience, and reliability. Preferred Experience Please mention in your application if you have experience with any of the following: CVAT Label Studio Labelbox Supervisely Roboflow V7 Bounding box annotation Polygon annotation Segmentation masks Keypoint labeling Object tracking Multi-frame video annotation Human activity or behavior labeling CCTV, retail, restaurant, warehouse, logistics, or industrial video datasets QA review for annotation projects Machine learning or computer vision data preparation Important Work Expectations This role may be structured as hourly or task-based depending on experience and project fit. You should be comfortable with: Remote work Clear annotation guidelines Daily or weekly throughput targets QA review of submitted work Rework when needed Feedback-driven improvement Maintaining consistency across many similar examples Asking good questions when instructions are unclear Quality matters more than speed, but both matter. We are looking for people who can produce accurate annotations at a reliable pace. How Your Work Will Be Evaluated Your performance will be measured using criteria such as: Annotation accuracy Agreement with QA review Consistency across similar cases Rework rate Frames/images completed per hour Timely completion of assigned batches Ability to follow project-specific labeling rules Why This Project Matters Your work will help train machine learning models that turn video footage into real operational intelligence. High-quality labeling will help improve model accuracy, reduce false detections, and support automation for real-world business environments. This is a foundational role in building reliable AI systems. The better the dataset, the better the model. To Apply Please include the following in your response: Your experience with image or video annotation The annotation tools you have used Whether you have worked with CCTV, retail, restaurant, warehouse, or operational video data Your availability per week Your hourly rate or task-based pricing preference A brief explanation of how you maintain accuracy when working on repetitive labeling tasks Whether you are willing to complete a small paid test task Applications that are generic or clearly copied and pasted will not be considered. Please start your response with: “CCTV Annotation – I understand the accuracy requirement.” This helps us confirm that you read the full posting. Strong performers may be offered ongoing work.