Key Responsibilities

• Annotate images and video content using object detection techniques (e.g., bounding boxes, segmentation).
• Identify and extract relevant frames from long-form video recordings (e.g., 1-hour videos).
• Label video actions and activities based on predefined categories.
• Review and validate annotations produced by the team to ensure consistency and accuracy.

Scope of Work

• Data Annotation:
• Annotate approximately 1 million images across 23 use cases.
• Draw accurate bounding boxes and assign correct class labels to each object.
• Maintain consistency and high standards in labeling.
• Target: 1000 annotated images per day per annotator.
• Quality Assurance:
• Conduct self-reviews before submission.
• Revise any flagged or substandard annotations.
• Data Management:
• Track progress using project management and data labeling platforms.
• Log any issues or ambiguities encountered during annotation tasks.

Requirements

Required Skills and Qualifications

• Proven experience in image and video annotation for object detection and action recognition.
• Ability to work independently, manage time efficiently, and meet tight deadlines.
• Basic understanding of computer vision concepts is advantageous.
• Familiarity with standard annotation tools such as CVAT, Labelbox, and Roboflow.