Depth information based separation of moving speakers' voices from mixed recordings.
Takahiro KigawaTaisuke SakanoHiroshi MizoguchiEtsuji YamaguchiShigenori InagakiYoshiaki TakedaMiki NamatameFusako KusunokiMasanori SugimotoRyohei EgusaPublished in: SMC (2016)
Keyphrases
- depth information
- depth map
- stereo vision
- depth recovery
- image based rendering
- depth images
- stereo camera
- speech recognition
- depth cameras
- depth data
- disparity map
- real time
- multi view
- rgbd images
- d scene
- time of flight
- stereoscopic video
- moving objects
- view interpolation
- shape from focus
- depth sensors
- depth estimation
- post processing
- image data
- hidden markov models
- high resolution
- disparity estimation
- viewpoint
- training data