Deep Visual-Semantic Hashing for Cross-Modal Retrieval
Yue Cao, Tsinghua university; Mingsheng Long*, Tsinghua University; Jianmin Wang, Tsinghua University; Qiang Yang, HKUST; Philip Yu, UIC
Due to the storage and retrieval eﬃciency, hashing has been widely applied to approximate nearest neighbor search for large-scale multimedia retrieval. Cross-modal hashing, which enables eﬃcient retrieval of images in response to text queries or vice versa, has received increasing attention recently. Most existing work on cross-modal hashing does not capture the spatial dependency of images and temporal dynamics of text sentences for learning powerful feature representations and cross-modal embeddings that mitigate the heterogeneity of diﬀerent modalities. This paper presents a new Deep Visual-Semantic Hashing (DVSH) model that generates compact hash codes of images and sentences in an end-to-end deep learning architecture, which capture the intrinsic cross-modal correspondences between visual data and natural language. DVSH is a hybrid deep architecture that constitutes a visual-semantic fusion network for learning joint embedding space of images and text sentences, and two modality-speciﬁc hashing networks for learning hash functions to generate compact binary codes. Our architecture eﬀectively uniﬁes joint multi-modal embedding and cross-modal hashing, which is based on a novel combination of Convolutional Neural Networks over images, Recurrent Neural Networks over sentences, and a structured max-margin objective that integrates all things together to enable learning of similarity-preserving and high-quality hash codes. Extensive empirical evidence shows that our DVSH approach yields state of the art results in cross-modal retrieval experiments on image-sentences datasets, i.e. standard IAPR TC-12 and large-scale Microsoft COCO.
Filed under: Big Data