Edge-guided Multi-domain RGB-to-TIR image Translation for Training Vision Tasks with Challenging Labels.
Dong-Guw LeeMyung-Hwan JeonYounggun ChoAyoung KimPublished in: ICRA (2023)
Keyphrases
- multi domain
- edge detection
- edge map
- input image
- image classification
- image features
- image retrieval
- multiscale
- labeled images
- computer vision
- pairwise
- ground truth labels
- similarity measure
- image segmentation
- image processing
- pixel values
- supervised learning
- gray scale
- training examples
- support vector machine
- training set
- real world
- cross domain
- edge information
- active learning
- color components
- search computing
- data mining