Bio-inspired computer vision: Towards a synergistic approach of artificial and biological vision
详细信息 查看全文
- 作者：N. V. Kartheek Medathati ; a ; kartheek.medathati@inria.fr" class="auth_mail" title="E-mail the corresponding author ; Heiko Neumannb ; heiko.neumann@uni-ulm.de" class="auth_mail" title="E-mail the corresponding author ; Guillaume S. Massonc ; guillaume.masson@univ-amu.fr" class="auth_mail" title="E-mail the corresponding author ; Pierre Kornprobsta ; pierre.kornprobst@inria.fr" class="auth_mail" title="E-mail the corresponding author
- 关键词：Canonical computations ; Event based processing ; Dynamic sensors ; Multiplexed representation ; Population coding ; Soft selectivity ; Feedback ; Lateral interactions ; Form-motion interactions
- 刊名：Computer Vision and Image Understanding
- 出版年：2016
- 出版时间：September 2016
- 年：2016
- 卷：150
- 期：Complete
- 页码：1-30
- 全文大小：3491 K
文摘
- class="listitem" id="list_celist0001">
- class="label">•
We present an overview of computational approaches to biological vision.
- class="label">•
We develop a task centered presentation of biological vision studies.
- class="label">•
We revisit three tasks: image sensing, scene segmentation and optical flow.
- class="label">•
We show how new computer vision methods could be developed from biological insights.
- class="label">•
We identify key task specific biological vision models which could be scaled up.