Seattle-based startup DataWeave, a competitive intelligence service provider for retailers and consumer brands, recently launched a counterfeit product detection system that uses deep learning to detect and help eliminate fake products from e-commerce websites.
“As online retailers and marketplaces aggressively add thousands of merchants on their platforms, unauthorized white labeling, listing fake products, and image theft have become persistent causes for concern,” Karthik Bettadapura, Co-founder, and CEO at DataWeave said.
Using NVIDIA GeForce GTX 1080 and GeForce GTX 1080 Ti GPUs with the cuDNN-accelerated TensorFlow and Caffe deep learning frameworks, the team trained their neural network on millions of catalog images of products such as electronics, cosmetics, apparel, footwear, and furniture.
“For certain categories like cosmetics, counterfeits can even be harmful to the consumer,” the company stated.
The system which utilizes the same GPUs used during training for inference has an accuracy of over 95% in detecting fake products.
The company says the fake products are spotted by “minute discrepancies” in catalog images and content, which the deep learning system can quickly detect.
“Our image analytics platform is highly advanced, capable of identifying and interpreting complex patterns and attributes in images of clothing and fashion accessories. Our machines recognize various fashion attributes by processing both images and associated text-based information available for a product,” Anshul Garg, the company’s AI specialist, said in a Medium post.
The product is available for retailers and e-commerce websites now.