Retrieving Similar E-Commerce Images Using Deep Learning
This work addresses the challenge of fine-grained image similarity for e-commerce applications, offering an incremental improvement over traditional deep CNNs.
The authors tackled the problem of retrieving visually similar e-commerce images by proposing a deep siamese convolutional neural network with a novel angular loss and combined embeddings, achieving superior image retrieval performance on four datasets compared to existing deep architectures.
In this paper, we propose a deep convolutional neural network for learning the embeddings of images in order to capture the notion of visual similarity. We present a deep siamese architecture that when trained on positive and negative pairs of images learn an embedding that accurately approximates the ranking of images in order of visual similarity notion. We also implement a novel loss calculation method using an angular loss metrics based on the problems requirement. The final embedding of the image is combined representation of the lower and top-level embeddings. We used fractional distance matrix to calculate the distance between the learned embeddings in n-dimensional space. In the end, we compare our architecture with other existing deep architecture and go on to demonstrate the superiority of our solution in terms of image retrieval by testing the architecture on four datasets. We also show how our suggested network is better than the other traditional deep CNNs used for capturing fine-grained image similarities by learning an optimum embedding.