Graph-RISE: Graph-Regularized Image Semantic Embedding
This work addresses the challenge of capturing nuanced image semantics for improved image retrieval and clustering, representing a strong specific gain in the domain of computer vision.
The paper tackles the problem of learning fine-grained image semantics for applications like search and clustering, and presents Graph-RISE, a neural graph learning framework that trains embeddings to discriminate 40M labels and outperforms state-of-the-art methods on tasks such as image classification and triplet ranking.
Learning image representations to capture fine-grained semantics has been a challenging and important task enabling many applications such as image search and clustering. In this paper, we present Graph-Regularized Image Semantic Embedding (Graph-RISE), a large-scale neural graph learning framework that allows us to train embeddings to discriminate an unprecedented O(40M) ultra-fine-grained semantic labels. Graph-RISE outperforms state-of-the-art image embedding algorithms on several evaluation tasks, including image classification and triplet ranking. We provide case studies to demonstrate that, qualitatively, image retrieval based on Graph-RISE effectively captures semantics and, compared to the state-of-the-art, differentiates nuances at levels that are closer to human-perception.