Evaluating Multimodal Representations on Visual Semantic Textual Similarity.
Oier Lopez de LacalleAnder SalaberriaAitor SoroaGorka AzkuneEneko AgirrePublished in: ECAI (2020)
Keyphrases
- visual representations
- semantic similarity
- visual information
- semantic representations
- visual similarity
- natural language
- semantic labels
- semantic content
- semantic similarity measure
- cross modal
- semantic context
- visual representation
- high level
- semantic information
- similarity measure
- mid level
- multi modal
- visual content
- multimedia
- semantic distance
- word similarity
- sentence similarity
- visual features
- higher level
- semantically relevant
- structural similarity
- intermediate representations
- low level
- word pairs
- audio visual
- metadata
- wordnet
- semantic web
- co occurrence
- multimodal interaction
- multimodal information
- visual and textual features
- textual descriptions
- visual perception
- textual information
- visual cues
- domain ontology
- semantic annotation
- visual data
- semantic network
- low level features
- similarity function
- keywords
- information extraction
- distance measure
- euclidean distance
- semantic concepts
- music retrieval
- semantic space
- semantic description
- semantic features
- visualization tools