Advancing Weakly-Supervised Audio-Visual Video Parsing via Segment-wise Pseudo Labeling.
Jinxing ZhouDan GuoYiran ZhongMeng WangPublished in: CoRR (2024)
Keyphrases
- audio visual
- weakly supervised
- visual data
- multimedia
- multi modal
- visual information
- video sequences
- video data
- object class
- relation extraction
- topic models
- superpixels
- semi supervised
- object detectors
- active learning
- multimedia data
- image segmentation
- natural language processing
- visual features
- video frames
- data sets
- image sequences
- image data
- probabilistic model
- pairwise
- object recognition