FlowGrad: Using Motion for Visual Sound Source Localization.
Rajsuryan SinghPablo ZinemanasXavier SerraJuan Pablo BelloMagdalena FuentesPublished in: CoRR (2022)
Keyphrases
- source localization
- sound source
- visual perception
- visual cues
- visual motion
- wireless sensor networks
- visual data
- visual information
- audio visual
- motion estimation
- image sequences
- low level
- multi modal
- camera motion
- visual features
- moving objects
- computational auditory scene analysis
- space time
- ground plane
- image processing
- human computer interaction
- image retrieval
- high level
- three dimensional