mlsanityData to 14 Oct 2026

Region-Aware Deformable Convolutions

arXiv 2509.15436 ↗First posted 18 Sep 2025cs.CV cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

We introduce Region-Aware Deformable Convolution (RAD-Conv), a new convolutional operator that enhances neural networks' ability to adapt to complex image structures. Unlike traditional deformable convolutions, which are limited to fixed quadrilateral sampling areas, RAD-Conv uses four boundary offsets per kernel element to create flexible, rectangular regions that dynamically adjust their size and shape to match image content. This approach allows precise control over the receptive field's width and height, enabling the capture of both local details and long-range dependencies, even with small 1x1 kernels. By decoupling the receptive field's shape from the kernel's structure, RAD-Conv combines the adaptability of attention mechanisms with the efficiency of standard convolutions. This innovative design offers a practical solution for building more expressive and efficient vision models, bridging the gap between rigid convolutional architectures and computationally costly attention-based methods.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 54 of its 61 references that are in the map. First posted on arXiv 18 Sep 2025.

Builds on

54 of 61 references.

References · 54

  1. 2017Attention Is All You Needcited by 60,361
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2021Swin Transformer: Hierarchical Vision Transformer using Shifted Windowscited by 12,139
  4. 2015Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networkscited by 9,639
  5. 2023Segment Anythingcited by 8,234
  6. 2020End-to-End Object Detection with Transformerscited by 7,414
  7. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,795
  8. 2017MobileNets: Efficient Convolutional Neural Networks for Mobile Vision Applicationscited by 3,845
  9. 2022A ConvNet for the 2020scited by 3,579
  10. 2020Deformable DETR: Deformable Transformers for End-to-End Object Detectioncited by 3,422
  11. 2015SSD: Single Shot MultiBox Detectorcited by 3,130
  12. 2021SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformerscited by 2,940
38 more
  1. 2021BEiT: BERT Pre-Training of Image Transformerscited by 2,375
  2. 2021Masked-attention Mask Transformer for Universal Image Segmentationcited by 2,351
  3. 2016Xception: Deep Learning with Depthwise Separable Convolutionscited by 2,332
  4. 2013Rich feature hierarchies for accurate object detection and semantic segmentationcited by 2,162
  5. 2021Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutionscited by 1,962
  6. 2019FCOS: Fully Convolutional One-Stage Object Detectioncited by 1,577
  7. 2017Deformable Convolutional Networkscited by 1,488
  8. 2022DINO: DETR with Improved DeNoising Anchor Boxes for End-to-End Object Detectioncited by 1,319
  9. 2022CoCa: Contrastive Captioners are Image-Text Foundation Modelscited by 1,274
  10. 2021Swin Transformer V2: Scaling Up Capacity and Resolutioncited by 1,228
  11. 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,133
  12. 2021Segmenter: Transformer for Semantic Segmentationcited by 855
  13. 2021CrossViT: Cross-Attention Multi-Scale Vision Transformer for Image Classificationcited by 806
  14. 2018CornerNet: Detecting Objects as Paired Keypointscited by 763
  15. 2021PVT v2: Improved Baselines with Pyramid Vision Transformercited by 751
  16. 2023Scaling Vision Transformers to 22 Billion Parameterscited by 694
  17. 2018Deformable ConvNets v2: More Deformable, Better Resultscited by 670
  18. 2019CenterNet: Keypoint Triplets for Object Detectioncited by 654
  19. 2014Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFscited by 554
  20. 2019Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selectioncited by 540
  21. 2023DETRs Beat YOLOs on Real-time Object Detectioncited by 533
  22. 2021CSWin Transformer: A General Vision Transformer Backbone with Cross-Shaped Windowscited by 527
  23. 2022InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutionscited by 482
  24. 2018Efficient Attention: Attention with Linear Complexitiescited by 368
  25. 2022Vision Transformer with Deformable Attentioncited by 295
  26. 2021Focal Self-attention for Local-Global Interactions in Vision Transformerscited by 288
  27. 2019Local Relation Networks for Image Recognitioncited by 210
  28. 2018Acquisition of Localization Confidence for Accurate Object Detectioncited by 176
  29. 2022More ConvNets in the 2020s: Scaling up Kernels Beyond 51x51 using Sparsitycited by 142
  30. 2024ViT-CoMer: Vision Transformer with Convolutional Multi-scale Feature Interaction for Dense Predictionscited by 66
  31. 2023UniRepLKNet: A Universal Perception Large-Kernel ConvNet for Audio, Video, Point Cloud, Time-Series and Image Recognitioncited by 64
  32. 2025Your ViT is Secretly an Image Segmentation Modelcited by 63
  33. 2025LLMDet: Learning Strong Open-Vocabulary Object Detectors under the Supervision of Large Language Modelscited by 57
  34. 2020Dynamic Region-Aware Convolutioncited by 50
  35. 2024Efficient Deformable ConvNets: Rethinking Dynamic and Sparse Operator for Vision Applicationscited by 50
  36. 2013Scalable Object Detection using Deep Neural Networkscited by 31
  37. 2024Breaking the Low-Rank Dilemma of Linear Attentioncited by 31
  38. 2025Vision Transformers on the Edge: A Comprehensive Survey of Model Compression and Acceleration Strategiescited by 24

50 most cited of 54 All 54 (JSON)

Built on by

Not cited yet.