Flipkart GRiD Challenge - Object Detection & Bounding Box Prediction
Efficient e-commerce product classification and object detection for the Flipkart GRiD Challenge.
This project was developed as part of the 7-day Flipkart Machine Learning Challenge and focuses on classifying e-commerce products into different categories while also implementing bounding-box object detection to identify the primary product within an image. This project addresses the task of efficient object detection in e-commerce product images without the use of pre-trained models. Leveraging standard architectures and a dataset comprising 24,000 training images and 24,045 testing images, the project achieved competitive results in bounding-box object detection, demonstrating the potential for robust e-commerce product localization in images.
If you have any questions or inquiries, please feel free to contact me at sahilvora2021@gmail.com