Name
Abstract | ||
|---|---|---|
Popular sites like Houzz, Pinterest, and LikeThatDecor, have communities of users helping each other answer questions about products in images. In this paper we learn an embedding for visual search in interior design. Our embedding contains two different domains of product images: products cropped from internet scenes, and products in their iconic form. With such a multi-domain embedding, we demonstrate several applications of visual search including identifying products in scenes and finding stylistically similar products. To obtain the embedding, we train a convolutional neural network on pairs of images. We explore several training architectures including re-purposing object classifiers, using siamese networks, and using multitask learning. We evaluate our search quantitatively and qualitatively and demonstrate high quality results for search across multiple visual domains, enabling new applications in interior design. |
Year | DOI | Venue |
|---|---|---|
2015 | 10.1145/2766959 | ACM Transactions on Graphics |
Keywords | Field | DocType |
visual similarity,interior design,deep learning,search | Visual search,Computer graphics (images),Computer science,Convolutional neural network,Artificial intelligence,Deep learning,Product design,The Internet,Computer vision,Multi-task learning,Embedding,Interior design,Machine learning | Journal |
Volume | Issue | ISSN |
34 | 4 | 0730-0301 |
Citations | PageRank | References |
109 | 3.13 | 26 |
Authors | ||
2 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Sean Bell | 1 | 448 | 19.59 |
| Kavita Bala | 2 | 2046 | 138.75 |