{"url":"/method/deformable-convolution","slug":"deformable-convolution","name":"Deformable Convolution","full_name":"Deformable Convolution","full_name_withheld":false,"description_markdown":"**Deformable convolutions** add 2D offsets to the regular grid sampling locations in the standard [convolution](https://paperswithcode.com/method/convolution). It enables free form deformation of the sampling grid. The offsets are learned from the preceding feature maps, via additional convolutional layers. Thus, the deformation is conditioned on the input features in a local, dense, and adaptive manner.","description_state":"present","introduced_year":null,"introduced_by":{"title":"Deformable Convolutional Networks","paper":"/paper/deformable-convolutional-networks","first_author":"Jifeng Dai","n_authors":7,"url_abs":null,"archive_paper_url":"https://paperswithcode.com/paper/deformable-convolutional-networks"},"source":{"url":"http://arxiv.org/abs/1703.06211v3","title":"Deformable Convolutional Networks","url_on_a_paper_host":true},"code_snippet_url":"https://github.com/chengdazhi/Deformable-Convolution-V2-PyTorch/blob/2f57c5db49161bd6c899670a5e4fba50e6b8fd26/modules/deform_conv.py#L10","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":152,"archive_num_papers":152,"papers_newest_first":[{"paper":null,"title":"MobileHolo: A Lightweight Complex-Valued Deformable CNN for High-Quality Computer-Generated Hologram","date":"2025-06-17","arxiv_id":"2506.14542","n_code_links":0,"syntology":null},{"paper":null,"title":"Detection of Underwater Multi-Targets Based on Self-Supervised Learning and Deformable Path Aggregation Feature Pyramid Network","date":"2025-05-21","arxiv_id":"2505.15518","n_code_links":0,"syntology":null},{"paper":null,"title":"DPN-GAN: Inducing Periodic Activations in Generative Adversarial Networks for High-Fidelity Audio Synthesis","date":"2025-05-14","arxiv_id":"2505.09091","n_code_links":0,"syntology":null},{"paper":null,"title":"An Appearance Defect Detection Method for Cigarettes Based on C-CenterNet","date":"2025-02-10","arxiv_id":"2502.06119","n_code_links":0,"syntology":null},{"paper":null,"title":"Low Resource Video Super-resolution using Memory and Residual Deformable Convolutions","date":"2025-02-03","arxiv_id":"2502.01816","n_code_links":0,"syntology":null},{"paper":"/paper/dstigcn-deformable-spatial-temporal","title":"DSTIGCN: Deformable Spatial-Temporal Interaction Graph Convolution Network for Pedestrian Trajectory Prediction","date":"2025-01-16","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":"/paper/a-universal-scale-adaptive-deformable","title":"A Universal Scale-Adaptive Deformable Transformer for Image Restoration across Diverse Artifacts","date":"2025-01-01","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":"/paper/model-decides-how-to-tokenize-adaptive-dna","title":"Model Decides How to Tokenize: Adaptive DNA Sequence Tokenization with MxDNA","date":"2024-12-18","arxiv_id":"2412.13716","n_code_links":1,"syntology":{"ran":2,"of":4,"unverified":2,"pointer_only":2}},{"paper":null,"title":"Towards 3D Semantic Scene Completion for Autonomous Driving: A Meta-Learning Framework Empowered by Deformable Large-Kernel Attention and Mamba Model","date":"2024-11-06","arxiv_id":"2411.03672","n_code_links":0,"syntology":null},{"paper":null,"title":"FlowDCN: Exploring DCN-like Architectures for Fast Image Generation with Arbitrary Resolution","date":"2024-10-30","arxiv_id":"2410.22655","n_code_links":0,"syntology":null},{"paper":"/paper/kaldex-kalman-filter-based-linear-deformable","title":"KaLDeX: Kalman Filter based Linear Deformable Cross Attention for Retina Vessel Segmentation","date":"2024-10-28","arxiv_id":"2410.21160","n_code_links":1,"syntology":null},{"paper":null,"title":"Mixture of Scale Experts for Alignment-free RGBT Video Object Detection and A Unified Benchmark","date":"2024-10-16","arxiv_id":"2410.12143","n_code_links":0,"syntology":null},{"paper":"/paper/optimizing-medical-image-segmentation-with","title":"Optimizing Medical Image Segmentation with Advanced Decoder Design","date":"2024-10-05","arxiv_id":"2410.04128","n_code_links":1,"syntology":null},{"paper":null,"title":"HRVMamba: High-Resolution Visual State Space Model for Dense Prediction","date":"2024-10-04","arxiv_id":"2410.03174","n_code_links":0,"syntology":null},{"paper":"/paper/gmt-enhancing-generalizable-neural-rendering","title":"GMT: Enhancing Generalizable Neural Rendering via Geometry-Driven Multi-Reference Texture Transfer","date":"2024-10-01","arxiv_id":"2410.00672","n_code_links":1,"syntology":null},{"paper":"/paper/burstm-deep-burst-multi-scale-sr-using","title":"BurstM: Deep Burst Multi-scale SR using Fourier Space with Optical Flow","date":"2024-09-21","arxiv_id":"2409.15384","n_code_links":1,"syntology":null},{"paper":null,"title":"KLDD: Kalman Filter based Linear Deformable Diffusion Model in Retinal Image Segmentation","date":"2024-09-19","arxiv_id":"2410.02808","n_code_links":0,"syntology":null},{"paper":null,"title":"DDNet: Deformable Convolution and Dense FPN for Surface Defect Detection in Recycled Books","date":"2024-09-08","arxiv_id":"2409.04958","n_code_links":0,"syntology":null},{"paper":null,"title":"MambaOcc: Visual State Space Model for BEV-based Occupancy Prediction with Local Adaptive Reordering","date":"2024-08-21","arxiv_id":"2408.11464","n_code_links":0,"syntology":null},{"paper":"/paper/u-decn-end-to-end-underwater-object-detection","title":"U-DECN: End-to-End Underwater Object Detection ConvNet with Improved DeNoising Training","date":"2024-08-11","arxiv_id":"2408.05780","n_code_links":1,"syntology":null},{"paper":null,"title":"Deformable Convolution Based Road Scene Semantic Segmentation of Fisheye Images in Autonomous Driving","date":"2024-07-23","arxiv_id":"2407.16647","n_code_links":0,"syntology":null},{"paper":null,"title":"SACNet: A Spatially Adaptive Convolution Network for 2D Multi-organ Medical Segmentation","date":"2024-07-14","arxiv_id":"2407.10157","n_code_links":0,"syntology":null},{"paper":null,"title":"Deformable Feature Alignment and Refinement for Moving Infrared Dim-small Target Detection","date":"2024-07-10","arxiv_id":"2407.07289","n_code_links":0,"syntology":null},{"paper":"/paper/edge-guided-and-cross-scale-feature-fusion","title":"Edge-guided and Cross-scale Feature Fusion Network for Efficient Multi-contrast MRI Super-Resolution","date":"2024-07-07","arxiv_id":"2407.05307","n_code_links":1,"syntology":null},{"paper":"/paper/sali-short-term-alignment-and-long-term-1","title":"SALI: Short-term Alignment and Long-term Interaction Network for Colonoscopy Video Polyp Segmentation","date":"2024-06-19","arxiv_id":"2406.13532","n_code_links":1,"syntology":null},{"paper":null,"title":"Coarse-Fine Spectral-Aware Deformable Convolution For Hyperspectral Image Reconstruction","date":"2024-06-18","arxiv_id":"2406.12703","n_code_links":0,"syntology":null},{"paper":"/paper/dehazedct-towards-effective-non-homogeneous","title":"DehazeDCT: Towards Effective Non-Homogeneous Dehazing via Deformable Convolutional Transformer","date":"2024-06-12","arxiv_id":null,"n_code_links":1,"syntology":null},{"paper":null,"title":"Adaptive Convolutional Forecasting Network Based on Time Series Feature-Driven","date":"2024-05-20","arxiv_id":"2405.12038","n_code_links":0,"syntology":null},{"paper":"/paper/a-new-multi-picture-architecture-for-learned","title":"A New Multi-Picture Architecture for Learned Video Deinterlacing and Demosaicing with Parallel Deformable Convolution and Self-Attention Blocks","date":"2024-04-19","arxiv_id":"2404.13018","n_code_links":1,"syntology":null},{"paper":"/paper/yolc-you-only-look-clusters-for-tiny-object","title":"YOLC: You Only Look Clusters for Tiny Object Detection in Aerial Images","date":"2024-04-09","arxiv_id":"2404.06180","n_code_links":1,"syntology":null}],"papers_shown":30,"tasks":[{"task":"/task/semantic-segmentation","name":"Semantic Segmentation","papers":30},{"task":"/task/object-detection","name":"Object Detection","papers":29},{"task":"/task/object-detection-1","name":"object-detection","papers":22},{"task":"/task/object","name":"Object","papers":20},{"task":"/task/segmentation","name":"Segmentation","papers":18},{"task":"/task/super-resolution","name":"Super-Resolution","papers":18},{"task":"/task/optical-flow-estimation","name":"Optical Flow Estimation","papers":12},{"task":"/task/image-segmentation","name":"Image Segmentation","papers":10},{"task":"/task/video-super-resolution","name":"Video Super-Resolution","papers":10},{"task":"/task/instance-segmentation","name":"Instance Segmentation","papers":9},{"task":"/task/decoder","name":"Decoder","papers":7},{"task":"/task/medical-image-segmentation","name":"Medical Image Segmentation","papers":7},{"task":"/task/autonomous-driving","name":"Autonomous Driving","papers":6},{"task":"/task/image-classification","name":"Image Classification","papers":6},{"task":"/task/deblurring","name":"Deblurring","papers":5},{"task":"/task/denoising","name":"Denoising","papers":5},{"task":"/task/image-classification","name":"image-classification","papers":5},{"task":"/task/image-generation","name":"Image Generation","papers":4},{"task":"/task/image-reconstruction","name":"Image Reconstruction","papers":4},{"task":"/task/motion-compensation","name":"Motion Compensation","papers":4}],"tasks_shown":20,"n_tasks":163,"usage_by_year":[{"year":"2017","papers":4},{"year":"2018","papers":7},{"year":"2019","papers":14},{"year":"2020","papers":30},{"year":"2021","papers":20},{"year":"2022","papers":24},{"year":"2023","papers":16},{"year":"2024","papers":30},{"year":"2025","papers":7}],"row_source":"methods_table","archive":{"source":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","archive_url":"https://paperswithcode.com/method/deformable-convolution"},"syntology_read_at":"2026-09-24T18:15:14+00:00"}