Self-supervised Semantic Segmentation Grounded in Visual Concepts.
Wenbin HeWilliam SurmeierArvind Kumar ShekarLiang GouLiu RenPublished in: IJCAI (2022)
Keyphrases
- visual concepts
- object categories
- pascal voc
- superpixels
- object detection
- object classes
- object recognition
- image patches
- training examples
- image content
- image collections
- image features
- semantic concepts
- scene classification
- visual information
- generative model
- image annotation
- natural images
- input image
- multiscale
- visual words
- multi modal
- visual features
- high level
- image database
- denoising
- image data
- prior knowledge
- image sequences