Anytime 3D Object Reconstruction Using Multi-Modal Variational Autoencoder.

Hyeonwoo Yu Jean Oh

Published in: IEEE Robotics Autom. Lett. (2022)

Keyphrases

multi modal
image segmentation
line drawings
reconstruction method
multi modality
high dimensional
video search
audio visual
cross modal
optical flow
machine learning
image processing
multiple modalities
semantic concepts
image annotation
markov random field
feature space
high level
feature selection