Publications

The idea of "learning to see like children" is described in the following arXiv paper (a reduced version was submitted for publication):
  • Gori, M., Lippi, M., Maggini, M., Melacci, S., Learning to see like children: proof of concept, arXiv:1408.2478 pdf

Some preliminary ideas can be found in the following publications:
  • Gori, M., Melacci, M., Lippi, M., Maggini, M., Information theoretic learning for pixel-based visual agents, European Conference on Computer Vision (ECCV), 2012. pdf
  • Melacci, M., Lippi, M., Gori, M., Maggini, M., Information-Based Learning of Deep Architectures for Feature Extraction, International Conference on Image Analysis and Processing (ICIAP), 2013. pdf
  • Frandina S., Lippi, M., Maggini, M., Melacci, M., On-Line Laplacian One-Class Support Vector Machines, International Conference on Artificial Neural Networks (ICANN), 2013. pdf
  • Gori, M., Lippi, M., Maggini, M., Melacci, S., On-line Video Motion Estimation by Invariant Receptive Inputs, Long-Term Detection and Tracking workshop at Computer Vision and Pattern Recognition (CVPR), IEEE Conference, 2014. pdf

Further information describing several aspects of Devlopmental Visual Agents can also be found in the following technical reports:
  • A survey on camera models and affine invariance. pdf
  • Some ongoing work regarding region aggregation and description. pdf