论文信息 - Smart Visual Sensing Using a Software Retina Model

Smart Visual Sensing Using a Software Retina Model

We present an approach to efficient visual sensing and perception based on a non-uniformly sampled, biologically inspired, software retina that when combined with a DCNN classifier has enabled megapixel-sized camera input images to be processed in a single pass, while maintaining state-of-the recognition performance.

Gerardo Aragon-Camarasa | Jan Paul Siebert | Piotr Ozimek | Lorinc Balog | Nina Hristozova

[1] E. L. Schwartz,et al. Spatial mapping in the primate sensory projection: Analytic structure and relevance to perception , 1977, Biological Cybernetics.

[2] Ian D. Reid,et al. Deep-6DPose: Recovering 6D Object Pose from a Single RGB Image , 2018, ArXiv.

[3] J. Paul Siebert,et al. Integrating a Non-Uniformly Sampled Software Retina with a Deep CNN Model , 2017 .

[4] R. Venkatesh Babu,et al. DeepFix: A Fully Convolutional Neural Network for Predicting Human Eye Fixations , 2015, IEEE Transactions on Image Processing.

[5] Jan Paul Siebert,et al. Efficient Egocentric Visual Perception Combining Eye-tracking, a Software Retina and Deep Learning , 2018, ArXiv.

[6] Sumitha L. Balasuriya. A computational model of space-variant vision based on a self-organised artificial retina tessellation , 2006 .

[7] Roland Wilson,et al. Self-similar Neural Networks Based on a Kohonen Learning Rule , 1996, Neural Networks.

[8] J. Paul Siebert,et al. Egocentric Perception using a Biologically Inspired Software Retina Integratedwith a Deep CNN , 2017 .

[9] Dima Damen,et al. Scaling Egocentric Vision: The EPIC-KITCHENS Dataset , 2018, ArXiv.

[10] Adam Schmidt,et al. A Software Retina for Egocentric & Robotic Vision Applications on Mobile Platforms , 2016 .

[11] Kai-Fu Yang,et al. Color Constancy Using Double-Opponency , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.