{"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/active-convolution-learning-the-shape-of","title":"Active Convolution: Learning the Shape of Convolution for Image Classification","arxiv_id":"1703.09076","date":"2017-03-27","proceeding":"CVPR 2017 7","authors":["Yunho Jeon","Junmo Kim"],"abstract":"In recent years, deep learning has achieved great success in many computer\nvision applications. Convolutional neural networks (CNNs) have lately emerged\nas a major approach to image classification. Most research on CNNs thus far has\nfocused on developing architectures such as the Inception and residual\nnetworks. The convolution layer is the core of the CNN, but few studies have\naddressed the convolution unit itself. In this paper, we introduce a\nconvolution unit called the active convolution unit (ACU). A new convolution\nhas no fixed shape, because of which we can define any form of convolution. Its\nshape can be learned through backpropagation during training. Our proposed unit\nhas a few advantages. First, the ACU is a generalization of convolution; it can\ndefine not only all conventional convolutions, but also convolutions with\nfractional pixel coordinates. We can freely change the shape of the\nconvolution, which provides greater freedom to form CNN structures. Second, the\nshape of the convolution is learned while training and there is no need to tune\nit by hand. Third, the ACU can learn better than a conventional unit, where we\nobtained the improvement simply by changing the conventional convolution to an\nACU. We tested our proposed method on plain and residual networks, and the\nresults showed significant improvement using our method on various datasets and\narchitectures in comparison with the baseline.","url_abs":"http://arxiv.org/abs/1703.09076v1","url_pdf":"http://arxiv.org/pdf/1703.09076v1.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":"active-convolution-learning-the-shape-of","repo_url":"https://github.com/jyh2986/Active-Convolution","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"image-classification","task_name":"image-classification"}],"methods":[{"method_slug":"1x1-convolution","method_name":"1x1 Convolution"},{"method_slug":"active-convolution","method_name":"Active Convolution"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"bottleneck-residual-block","method_name":"Bottleneck Residual Block"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"global-average-pooling","method_name":"Global Average Pooling"},{"method_slug":"grouped-convolution","method_name":"Grouped Convolution"},{"method_slug":"local-response-normalization","method_name":"Local Response Normalization"},{"method_slug":"max-pooling","method_name":"Max Pooling"},{"method_slug":"randomhorizontalflip","method_name":"Random Horizontal Flip"},{"method_slug":"random-resized-crop","method_name":"Random Resized Crop"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"sgd-with-momentum","method_name":"SGD with Momentum"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"step-decay","method_name":"Step Decay"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"zca-whitening","method_name":"ZCA Whitening"}],"datasets_introduced":[],"methods_introduced":[{"slug":"active-convolution","name":"Active Convolution","full_name":"Active Convolution"}],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.09076","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}