SSVOD: Semi-Supervised Video Object Detection with Sparse Annotations.
Tanvir MahmudChun-Hao LiuBurhaneddin YamanDiana MarculescuPublished in: CoRR (2023)
Keyphrases
- object detection
- semi supervised
- video data
- semi supervised learning
- video sequences
- natural language descriptions
- video content
- multi view
- high dimensional
- computer vision
- video streams
- multi class
- real time
- video material
- face detection
- video frames
- multimedia
- real time face tracking
- action detection
- human detection
- labeled data
- unlabeled data
- active learning
- sparse representation
- sparse coding
- pedestrian detection
- scene understanding
- semi supervised classification
- subspace learning
- user generated
- keywords
- background subtraction
- video clips
- video surveillance
- unsupervised learning
- semantic annotation
- event detection
- human activities