Fashion Captioning: Towards Generating Accurate Descriptions with Semantic Rewards.
Xuewen YangHeming ZhangDi JinYingru LiuChi-Hao WuJianchao TanDongliang XieJue WangXin WangPublished in: ECCV (13) (2020)
Keyphrases
- semantic description
- high level
- semantic representations
- natural language
- high accuracy
- semantic web
- markov decision processes
- semantic similarity
- reinforcement learning
- semantic level
- semantic representation
- bandit problems
- database
- semantic descriptions
- semantic knowledge
- semantic relationships
- context dependent
- highly accurate
- semantic network
- semantic annotation
- semantic information
- high quality
- information systems
- natural language descriptions
- genetic algorithm