FoRConvD: An approach for food recognition on mobile devices using convolutional neural networks and depth maps.
Vlad-Ioan TomescuPublished in: SACI (2020)
Keyphrases
- depth map
- convolutional neural networks
- mobile devices
- depth information
- multi view
- convolutional network
- d scene
- stereo matching
- three dimensional
- view synthesis
- high quality
- time of flight
- image sequences
- input image
- multibaseline stereo
- intermediate views
- object recognition
- depth cameras
- depth estimation
- high resolution
- real scenes
- depth images
- multi view stereo
- multiview images
- super resolution
- viewpoint
- feature extraction
- stereo vision
- computer vision
- free viewpoint
- stereoscopic images
- multiscale
- action recognition