Cycle-Correspondence Loss: Learning Dense View-Invariant Visual Features from Unlabeled and Unordered RGB Images.
David B. AdrianAndras Gabor KupcsikMarkus SpiesHeiko NeumannPublished in: ICRA (2024)
Keyphrases
- visual features
- image classification
- image collections
- image retrieval
- image annotation
- visual appearance
- web images
- visual content
- visual information
- labeled images
- learning algorithm
- supervised learning
- visual similarity
- low level
- input image
- image matching
- view invariant
- low level features
- image search
- human actions
- three dimensional
- visual data
- semantic gap
- image data
- active learning
- multi view
- image features
- bag of features
- low level visual features
- keywords
- image content
- keypoints
- semi supervised learning
- image database
- image registration
- semi supervised
- viewpoint