DiffuMask: Synthesizing Images with Pixel-level Annotations for Semantic Segmentation Using Diffusion Models.
Weijia WuYuzhong ZhaoMike Zheng ShouHong ZhouChunhua ShenPublished in: ICCV (2023)
Keyphrases
- pixel level
- superpixels
- semantic segmentation
- input image
- image fusion
- multiple images
- image data
- segmentation method
- multiscale
- image registration
- segmentation algorithm
- foreground objects
- bounding box
- ground truth
- three dimensional
- object class
- image features
- image annotation
- conditional random fields
- object classes
- object segmentation
- image understanding
- image patches
- test images
- object detection
- image analysis
- natural images