Recurrent Topic-Transition GAN for Visual Paragraph Generation
arXiv:1703.07022
Abstract
A natural image usually conveys rich semantic content and can be viewed from different angles. Existing image description methods are largely restricted by small sets of biased visual paragraph annotations, and fail to cover rich underlying semantics. In this paper, we investigate a semi-supervised paragraph generative framework that is able to synthesize diverse and semantically coherent paragraph descriptions by reasoning over local semantic regions and exploiting linguistic knowledge. The proposed Recurrent Topic-Transition Generative Adversarial Network (RTT-GAN) builds an adversarial framework between a structured paragraph generator and multi-level paragraph discriminators. The paragraph generator generates sentences recurrently by incorporating region-based visual and language attention mechanisms at each step. The quality of generated paragraph sentences is assessed by multi-level adversarial discriminators from two aspects, namely, plausibility at sentence level and topic-transition coherence at paragraph level. The joint adversarial training of RTT-GAN drives the model to generate realistic paragraphs with smooth logical transition between sentence topics. Extensive quantitative experiments on image and video paragraph datasets demonstrate the effectiveness of our RTT-GAN in both supervised and semi-supervised settings. Qualitative results on telling diverse stories for an image also verify the interpretability of RTT-GAN.
10 pages, 6 figures
References in corpus (6)
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Microsoft COCO Captions: Data Collection and Evaluation Server
- Deep Fragment Embeddings for Bidirectional Image Sentence Mapping
- Deep Captioning with Multimodal Recurrent Neural Networks (m-RNN)
- Deep Visual-Semantic Alignments for Generating Image Descriptions
- Deep Variation-structured Reinforcement Learning for Visual Relationship and Attribute Detection
Cited by in corpus (6)
- The Effectiveness of Data Augmentation in Image Classification using Deep Learning
- Language Generation with Recurrent Generative Adversarial Networks without Pre-training
- Structured Generative Adversarial Networks
- From Deterministic to Generative: Multi-Modal Stochastic RNNs for Video Captioning
- Dual Motion GAN for Future-Flow Embedded Video Prediction
- Efficient Correlated Topic Modeling with Topic Embedding