USD: Uncertainty-Based One-Phase Learning to Enhance Pseudo-Label Reliability for Semi-Supervised Object Detection.
Dayoung ChunSeungil LeeHyun KimPublished in: IEEE Trans. Multim. (2024)
Keyphrases
- object detection
- semi supervised
- learning phase
- active learning
- face detection
- labeled and unlabeled data
- supervised learning
- learning algorithm
- reinforcement learning
- multi class
- learning process
- computer vision
- learning systems
- learning problems
- neural network
- pedestrian detection
- uncertain data
- multi label learning
- conditional random fields
- background knowledge
- unlabeled data
- unsupervised learning
- knowledge acquisition
- online learning
- prior knowledge