{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/unsupervised-learning-of-a-hierarchical","title":"Unsupervised Learning of a Hierarchical Spiking Neural Network for Optical Flow Estimation: From Events to Global Motion Perception","arxiv_id":"1807.10936","date":"2018-07-28","proceeding":null,"authors":["Federico Paredes-Vallés","Kirk Y. W. Scheper","Guido C. H. E. de Croon"],"abstract":"The combination of spiking neural networks and event-based vision sensors\nholds the potential of highly efficient and high-bandwidth optical flow\nestimation. This paper presents the first hierarchical spiking architecture in\nwhich motion (direction and speed) selectivity emerges in an unsupervised\nfashion from the raw stimuli generated with an event-based camera. A novel\nadaptive neuron model and stable spike-timing-dependent plasticity formulation\nare at the core of this neural network governing its spike-based processing and\nlearning, respectively. After convergence, the neural architecture exhibits the\nmain properties of biological visual motion systems, namely feature extraction\nand local and global motion perception. Convolutional layers with input\nsynapses characterized by single and multiple transmission delays are employed\nfor feature and local motion perception, respectively; while global motion\nselectivity emerges in a final fully-connected layer. The proposed solution is\nvalidated using synthetic and real event sequences. Along with this paper, we\nprovide the cuSNN library, a framework that enables GPU-accelerated simulations\nof large-scale spiking neural networks. Source code and samples are available\nat https://github.com/tudelft/cuSNN.","url_abs":"http://arxiv.org/abs/1807.10936v2","url_pdf":"http://arxiv.org/pdf/1807.10936v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"unsupervised-learning-of-a-hierarchical","repo_url":"https://github.com/tudelft/cuSNN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"event-based-vision","task_name":"Event-based vision"},{"task_slug":null,"task_name":"GPU"},{"task_slug":"optical-flow-estimation","task_name":"Optical Flow Estimation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1807.10936","atlas_url":"https://app.syntology.ai/?focus=1807.10936","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}