SE-MD: a single-encoder multiple-decoder deep network for point cloud reconstruction from 2D images.
Abdul Mueed HafizRouf Ul Alam BhatShabir Ahmad ParahM. HassaballahPublished in: Pattern Anal. Appl. (2023)
Keyphrases
- point cloud
- low complexity
- structure from motion
- architectural scenes
- laser scanner
- video codec
- surface reconstruction
- dense reconstruction
- high resolution
- point sets
- multi view stereo
- image reconstruction
- decoding process
- implicit surfaces
- distributed video coding
- visual hull
- high dimensional
- missing data
- reconstructed surface
- knn
- urban scenes
- motion estimation
- reconstruction method
- stereo camera
- error concealment
- three dimensional
- nearest neighbor
- vision system
- bit rate
- hough transform