{"url":"/method/deformable-kernel","slug":"deformable-kernel","name":"Deformable Kernel","full_name":"Deformable Kernel","full_name_withheld":false,"description_markdown":"A **Deformable Kernels** is a type of convolutional operator for deformation modeling. DKs learn free-form offsets on kernel coordinates to deform the original kernel space towards specific data modality, rather than recomposing data. This can directly adapt the effective receptive field (ERF) while leaving the receptive field untouched. They can be used as a drop-in replacement of rigid kernels. \r\n\r\nAs shown in the Figure, for each input patch, a local DK first generates a group of kernel offsets $\\{\\Delta \\mathcal{k}\\}$ from input feature patch using the light-weight generator $\\mathcal{G}$ (a 3$\\times$3 [convolution](https://paperswithcode.com/method/convolution) of rigid kernel). Given the original kernel weights $\\mathcal{W}$ and the offset group $\\{\\Delta \\mathcal{k}\\}$, DK samples a new set of kernel $\\mathcal{W}'$ using a bilinear sampler $\\mathcal{B}$. Finally, DK convolves the input feature map and the sampled kernels to complete the whole computation.","description_state":"present","introduced_year":null,"introduced_by":{"title":null,"paper":null,"first_author":null,"n_authors":0,"url_abs":null,"archive_paper_url":null},"source":{"url":"https://arxiv.org/abs/1910.02940v2","title":"Deformable Kernels: Adapting Effective Receptive Fields for Object Deformation","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/hangg7/deformable-kernels/blob/886c35389ab9fb96b36d2ab89c56186381acff32/deformable_kernels/modules/deform_kernel.py#L228","code_snippet_url_on_a_code_host":true,"categories":[{"area":"Computer Vision","area_id":"computer-vision","collection":"Convolutions","url":"/methods/category/convolutions","pwc_aliases":[]}],"n_papers_tagged":8,"archive_num_papers":null,"papers_newest_first":[{"paper":null,"title":"RobuRCDet: Enhancing Robustness of Radar-Camera Fusion in Bird's Eye View for 3D Object Detection","date":"2025-02-18","arxiv_id":"2502.13071","n_code_links":0,"syntology":null},{"paper":"/paper/revisiting-deformable-convolution-for-depth","title":"Revisiting Deformable Convolution for Depth Completion","date":"2023-08-03","arxiv_id":"2308.01905","n_code_links":2,"syntology":null},{"paper":null,"title":"Deformable Kernel Expansion Model for Efficient Arbitrary-shaped Scene Text Detection","date":"2023-03-28","arxiv_id":"2303.15737","n_code_links":0,"syntology":null},{"paper":null,"title":"H-VFI: Hierarchical Frame Interpolation for Videos with Large Motions","date":"2022-11-21","arxiv_id":"2211.11309","n_code_links":0,"syntology":null},{"paper":null,"title":"FisheyeHDK: Hyperbolic Deformable Kernel Learning for Ultra-Wide Field-of-View Image Recognition","date":"2022-03-14","arxiv_id":"2203.07255","n_code_links":0,"syntology":null},{"paper":null,"title":"Deformable Kernel Convolutional Network for Video Extreme Super-Resolution","date":"2020-10-01","arxiv_id":"2010.00154","n_code_links":0,"syntology":null},{"paper":"/paper/deformable-kernel-networks-for-joint-image","title":"Deformable Kernel Networks for Joint Image Filtering","date":"2019-10-17","arxiv_id":"1910.08373","n_code_links":2,"syntology":null},{"paper":"/paper/deformable-kernels-adapting-effective","title":"Deformable Kernels: Adapting Effective Receptive Fields for Object Deformation","date":"2019-10-07","arxiv_id":"1910.02940","n_code_links":2,"syntology":null}],"papers_shown":8,"tasks":[{"task":"/task/object","name":"Object","papers":2},{"task":"/task/object-detection","name":"Object Detection","papers":2},{"task":"/task/3d-object-detection","name":"3D Object Detection","papers":1},{"task":"/task/depth-completion","name":"Depth Completion","papers":1},{"task":"/task/depth-map-super-resolution","name":"Depth Map Super-Resolution","papers":1},{"task":"/task/graph-matching","name":"Graph Matching","papers":1},{"task":"/task/image-classification","name":"Image Classification","papers":1},{"task":"/task/image-restoration","name":"Image Restoration","papers":1},{"task":"/task/object-recognition","name":"Object Recognition","papers":1},{"task":"/task/robust-object-detection","name":"Robust Object Detection","papers":1},{"task":"/task/scene-text-detection","name":"Scene Text Detection","papers":1},{"task":"/task/segmentation","name":"Segmentation","papers":1},{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":1},{"task":"/task/super-resolution","name":"Super-Resolution","papers":1},{"task":"/task/text-detection","name":"Text Detection","papers":1},{"task":"/task/video-frame-interpolation","name":"Video Frame Interpolation","papers":1},{"task":"/task/video-super-resolution","name":"Video Super-Resolution","papers":1},{"task":"/task/object-detection-1","name":"object-detection","papers":1}],"tasks_shown":18,"n_tasks":18,"usage_by_year":[{"year":"2019","papers":2},{"year":"2020","papers":1},{"year":"2022","papers":2},{"year":"2023","papers":2},{"year":"2025","papers":1}],"row_source":"embedded","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/deformable-kernel"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}