Adaptive Data Over-Compression Method for 3D Object Reconstruction using Multi-modal Variational Autoencoder

Hyeonwoo Yu, Jean Oh

Abstract

For effective human-robot teaming, it is important for the robots to be able to share their visual perception with the human operators. In a harsh remote collaboration setting, however, it is especially challenging to transfer a large amount of sensory data over a low-bandwidth network in real-time, e.g., for the task of 3D shape reconstruction given 2D camera images. To reduce the burden of data transferring, data compression techniques such as autoencoder can be utilized to obtain and transmit the data in terms of latent variables in a compact form. However, to overcome the lower-bandwidth limitation and achieve faster transmission, an adaptive and flexible over-compression method is necessary that can exploit only partial elements of the latent variables. To handle these cases, we propose a method for imputation of latent variables whose elements are partially excluded for additional compression. To perform imputation with only some dimensions of variables, exploiting prior information of the category- or instance-level is essential. In general, a prior distribution used in variational autoencoders is achieved from all of the training datapoints regardless of their labels. This type of flattened prior makes it difficult to perform imputation from the category- or instance-level distributions. We overcome this limitation by exploiting a category-specific multi-modal prior distribution in the latent space. By finding a modal in a latent space according to the remaining elements of the latent variables, the missing elements can be sampled. Based on the experiments on the ModelNet and Pascal3D datasets, the proposed approach shows a consistently superior performance over autoencoder and variational autoencoder up to 50\% data loss.
BibTeX
@article{yu2022adaptive,
  author = {Hyeonwoo Yu and Jean Oh},
  title = {{Adaptive Data Over-Compression Method for 3D Object Reconstruction using Multi-modal Variational Autoencoder}},
  journal = {IEEE Robotics Autom. Lett.},
  year = {2022},
  url = {https://openreview.net/forum?id=Z9MXMqsJrE},
}
Figure