🔬 Excited to present our latest research at the #MAIN2025 conference today!
👁️ What if CNNs could see like humans? Our new work shows how foveated vision—concentrating processing at gaze center—makes networks more robust to perturbations & great at localization. Inspired by human vision’s architecture (high-resolution foveal center, low-resolution periphery), we embedded this retinotopic transformation into CNN architectures, allowing to actively scan the image. This gives literally a new look to #ConvNets !
📄 Paper:
(2025).
Foveated Retinotopy Improves Classification and Localization in CNNs.
Submitted.
#DeepLearning #ComputerVision #AI #Research #NeuralNetworks #NeuroAI #OpenScience I love #Montreal
Media storm - share if you like :
- https://neuromatch.social/@laurentperrinet/115712116667035852
- https://www.linkedin.com/posts/laurent-perrinet-1857b9_main2025-convnets-deeplearning-activity-7405576210266890240-mOLV
- https://bsky.app/profile/laurentperrinet.bsky.social/post/3m7ukcmx4nk24
- https://www.instagram.com/reel/DSM-apSj1GH/
- https://www.facebook.com/share/v/17inxsTpwy/
- https://youtu.be/xnnwG0dkKNk
Laurent U Perrinet
Researcher in Computational Neuroscience
My research interests include Machine Learning and computational neuroscience applied to Vision.