Video Annotation for Intelligent Transport -- 3
Budget: ₹12,500 – ₹37,500 INR
I’m expanding our computer-vision pipeline for intelligent transportation and now need extra hands to turn raw driving footage into crisp, production-ready training data. Your main responsibilities will be twofold: first, frame-by-frame video annotation that accurately marks every vehicle on screen; second, independent validation passes to confirm label quality before the files move downstream to our model engineers.
All footage is supplied through our secure web platform, and I walk you through the entire workflow during a short, paid onboarding session—so no prior AI background is necessary. You work remotely on your own schedule, submit batches whenever they’re ready, and receive regular payouts tied to each approved milestone.
Core deliverables
• Precisely annotated video clips with bounding boxes (or polygons when required) around every vehicle
• A brief validation report per clip confirming object count, label consistency, and any edge-case notes
• Timely upload of the final reviewed dataset in the same directory structure we provide
We’re starting with vehicles only, but there’s room to branch into pedestrians and traffic signs as new projects roll out, so attention to detail and a willingness to follow evolving guidelines is key. Tools such as CVAT or Labelbox are integrated in the portal, but I’m open to suggestions if you have another favorite video annotation environment.
If you’re meticulous, comfortable working with video, and eager to contribute to real-world autonomous mobility, I’d love to bring you onto the team.
I have uploaded the guidelines. Please check and if you are okay with it, please contact us. So that we can give access to our CVAT portal as well as slack channel.
All footage is supplied through our secure web platform, and I walk you through the entire workflow during a short, paid onboarding session—so no prior AI background is necessary. You work remotely on your own schedule, submit batches whenever they’re ready, and receive regular payouts tied to each approved milestone.
Core deliverables
• Precisely annotated video clips with bounding boxes (or polygons when required) around every vehicle
• A brief validation report per clip confirming object count, label consistency, and any edge-case notes
• Timely upload of the final reviewed dataset in the same directory structure we provide
We’re starting with vehicles only, but there’s room to branch into pedestrians and traffic signs as new projects roll out, so attention to detail and a willingness to follow evolving guidelines is key. Tools such as CVAT or Labelbox are integrated in the portal, but I’m open to suggestions if you have another favorite video annotation environment.
If you’re meticulous, comfortable working with video, and eager to contribute to real-world autonomous mobility, I’d love to bring you onto the team.
I have uploaded the guidelines. Please check and if you are okay with it, please contact us. So that we can give access to our CVAT portal as well as slack channel.