Job Opportunity
As a video annotator, you will be responsible for accurately placing skeleton keypoints and drawing bounding boxes on video content. The ideal candidate should have strong skills in annotation placement using COCO or custom schema.
Key Responsibilities :
- Accurately place skeleton keypoints with high precision
- Ensure consistent joint alignment across frames, even in cases of motion blur or occlusion
- Draw precise and smooth bounding boxes across all frames
- Handle scale, perspective, and partially visible subjects effectively
Required Skills :
Proficient frame-by-frame labeling and interpolation abilitiesMaintain temporal consistency in skeletons and bounding boxesDetect and fix annotation drift in long sequencesBalance speed and accuracy in complex datasetsCognitive Abilities :
Excellent attention to detail and spatial awarenessStrong understanding of body posture and motionResolve ambiguous or unclear frames effectivelyConsistently apply labeling rulesQuality Assurance :
Familiarity with IoU (bounding boxes) and PCK (keypoints) metricsReview and correct annotations before submissionUnderstand how data quality affects AI / ML model performancePreferred Background :
Knowledge of basic anatomyExperience with pose estimation datasetsWorked on sports, gesture, or action-tracking projectsFamiliarity with platforms like V7 Darwin