Academic2024
Fashion Object Detection
CS271 AI Fundamentals: Object Detection for Automated Product Tagging
Role
AI Engineer
Year
2024
Team
Group Project
Tech Stack
YOLO, Roboflow, Python, GPU Cloud (Runpod.io)
A group project in CS271 (AI Fundamentals) focused on building a computer vision system that detects and extracts clothing attributes from images to automate product tagging for e-commerce. The project emphasizes practical tags for apparel—such as color, clothing type, and patterns—to reduce manual labeling effort and improve consistency across large product image catalogs.
01 The Problem
- Manual product tagging is time-consuming and error-prone, leading to inconsistent metadata.
- Large product image catalogs require automated extraction of key attributes for search and categorization.
- The solution needs a fast, practical detection approach suitable for real-world e-commerce workflows.
02 The Solution
- Owned dataset preparation and image labeling in Roboflow, defining target classes/attributes (e.g., color, clothing type, pattern) and ensuring consistent annotations.
- Adopted YOLO for object detection due to its speed and practicality for inference-oriented use cases.
- Managed training infrastructure on GPU Cloud (Runpod.io) to enable efficient training and iterative experimentation.
- Established a training and evaluation workflow, iterating on key settings to better fit real-world image variability.
03 The Result
- Delivered a prototype that detects clothing items and extracts key attributes for automated tagging (e.g., color, type, patterns).
- Reduced manual tagging effort while improving metadata consistency for e-commerce inventory search and management.
- Produced a repeatable pipeline—from dataset preparation to training and inference—that can be extended with additional classes or attributes.
Project Gallery
Inference example: detected items and extracted attributes
