Object Detection and Segmentation Auto-Labeling

Developed automatic and manual labeling pipeline for detection and segmentation datasets.

Personal ProjectsYear 2025Latest activity 2026-03-07

Project Overview

Objective

Developed automatic and manual labeling pipeline for detection and segmentation datasets.

Stack

GroundingDINOSAMYOLOv8YOLOv8-seg

Delivery highlights

Implemented automatic labeling with GroundingDINO for text-guided detection and SAM for segmentation masks, Converted generated masks to YOLO-format annotations for training YOLOv8 and YOLOv8-seg models, and Implemented manual labeling loop: seed annotation, model-assisted labeling, review, correction, and retraining.

Back to Topic Projects Back to All Projects

System workflow

How model-assisted auto-labeling works

A human-in-the-loop pipeline generates candidate annotations, validates them, trains a model, and feeds improvements into the next labeling cycle.

Rendering workflow diagram…

Scroll horizontally to explore the full workflow on smaller screens.

System workflow

Foundation model assisted labeling pipeline

GroundingDINO and SAM work together to generate automatic annotations from text prompts and images, then feed refined labels back to YOLO training.

Rendering workflow diagram…

Scroll horizontally to explore the full workflow on smaller screens.

System workflow

Model-assisted annotation workflow

Iterative cycle where improved models generate better candidate labels, reducing manual annotation effort with each iteration.

Rendering workflow diagram…

Scroll horizontally to explore the full workflow on smaller screens.

System workflow

Source-reviewed processing sequence

A conservative end-to-end boundary taken from the source audit. Specific endpoints, storage writes, and external calls are only claimed in the project-specific diagrams when the implementation confirms them.

Rendering workflow diagram…

Scroll horizontally to explore the full workflow on smaller screens.

Related Projects

3 items

Semi-Supervised Video Object Segmentation for Electric Vehicle Charger Socket Tracking

Personal ProjectsYear: 2026

Extended previous Electric Vehicle Charger Socket Instance Segmentation and Electric Vehicle Charger Socket Instance Segmentation & Tracking projects by building a custom DAVIS-style dataset with YOLOv8-Seg and adapting it for XMem to enable temporally consistent pixel-level segmentation across video frames.

Small-Object Focused Augmentation for Electric Vehicle Charger Socket Detection and Tracking

Personal ProjectsYear: 2026

Built upon the previous Semi-Supervised Video Object Segmentation for Electric Vehicle Charger Socket Tracking project by applying small-object-focused data augmentation to address missed detections when charger sockets appeared smaller or farther from the camera. This improved the model’s ability to detect small and hard-to-detect sockets while reducing missed detections in real-world scenarios.

Electric Vehicle Charger Socket Instant Segmentation

Personal ProjectsYear: 2025

Built and benchmarked EV charger socket instance segmentation models for real-time use.