D2DF2WOD: Learning Object Proposals for Weakly-Supervised Object Detection via Progressive Domain Adaptation.
Yuting WangRicardo GuerreroVladimir PavlovicPublished in: CoRR (2022)
Keyphrases
- domain adaptation
- learning objects
- weakly supervised
- object detection
- object class
- object detectors
- semi supervised
- superpixels
- object segmentation
- object categories
- cross domain
- multiple sources
- metadata
- labeled data
- object recognition
- multi class
- e learning
- computer vision
- semi supervised learning
- learning process
- unlabeled data
- transfer learning
- target domain
- topic models
- bounding box
- active learning
- co training
- training data
- databases
- supervised learning
- small number
- pairwise