Skip to content

AI Data Annotation: Dhaka Street Scene, Boxes and Person Polygons

A CC0 Dhaka street photo labeled for object detection: vehicles, rickshaws and a sign as boxes, people as anonymized silhouettes, exported to COCO JSON.

Before: AI Data Annotation: Dhaka Street Scene, Boxes and Person Polygons
After: AI Data Annotation: Dhaka Street Scene, Boxes and Person Polygons
Source photo (anonymized) Annotated, 6 classes

AI Data Annotation: Dhaka Street Scene, Boxes and Person Polygons

Manual bounding boxes and person polygons, occlusion and truncation attributes, face, plate and logo blurring, COCO JSON export

Studio demo on a public-domain photo, not a client job

Bounding Box Annotation AI & Computer Vision 24-48 hours

Technique
Manual bounding boxes and person polygons, occlusion and truncation attributes, face, plate and logo blurring, COCO JSON export
Industry
AI & Computer Vision
Turnaround
24-48 hours
This is a MediaScaleUp studio demonstration on a CC0 street photo near Dhaka from Wikimedia Commons. We blurred faces, the license plate and vehicle branding, then drew 15 annotations by hand across six classes: bounding boxes for the van, car, cycle rickshaws, cargo tricycle and street sign, and polygon silhouettes for people. The labels are delivered as a COCO JSON file alongside the image.

Try us free on your own work.

Send us something real and we will do it to your brief — no card, no commitment. If the standard is not what you need, you have lost nothing.