BMCR: Adaptive Backbone Module Composition via Reinforcement Learning for Remote Sensing Object Detection

• URL: http://arxiv.org/abs/2606.05586v1
• Date: Thu, 04 Jun 2026 02:04:15 GMT
• Title: BMCR: Adaptive Backbone Module Composition via Reinforcement Learning for Remote Sensing Object Detection
• Abstract summary: We propose Backbone Module Composition via Reinforcement Learning (BMCR) for object detection detectors.<n> BMCR assembles input-adaptive inference paths from reusable modules decomposed from off-the-shelf CNN and ViT backbones.<n>We show that BMCR achieves 79.31%, 73.41% and 71.86% mAP, respectively, surpassing strong static and dynamic baselines by up to 2.5 points.
• Score: 4.303688954031907
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In remote sensing object detection, Convolutional Neural Networks (CNNs) excel at capturing local details while Vision Transformers (ViTs) are better at global context modeling. However, existing detectors typically rely on a single fixed backbone or a manually designed hybrid architecture, and thus fail to adaptively exploit these complementary strengths across inputs of diverse complexity. To address this limitation, we propose Backbone Module Composition via Reinforcement Learning (BMCR). BMCR dynamically assembles input-adaptive inference paths from reusable modules decomposed from off-the-shelf CNN and ViT backbones. To enable such cross-family composition, we first construct an extensible module toolbox. Specifically, we decompose representative CNN and ViT backbones into reusable functional modules and encapsulate each module with explicit structural, semantic, and computational metadata for compatibility-aware assembly. To bridge the gap between grid-based CNN features and token-based ViT representations, we design a lightweight Optimal Transport (OT) based transition interface that ensures distribution-aware alignment while respecting spatial consistency. The backbone composition process is then formulated as a sequential decision problem, in which a policy network progressively selects task-relevant modules according to intermediate multi-scale observations. To stabilize the joint optimization of reusable modules and the routing policy, we further develop an Adaptive Module Cooperative Optimization (AMCO) strategy that coordinates module updating, routing exploration, and reward assignment during training. On DOTA-v1.0, DOTA-v1.5 and DIOR-R, BMCR achieves 79.31\%, 73.41\% and 71.86\% mAP, respectively, surpassing strong static and dynamic baselines by up to 2.5 points while maintaining competitive efficiency.

Related papers

• Adaptive Cross-Modal Fusion with Sparse Attention for Pedestrian Crossing Intention Prediction [4.8915598762204135]
  ADAPT (Adaptive Domain-Aware Pedestrian Crossing Transformer) is a multimodal framework for predicting pedestrian crossing intention.<n> ADAPT processes four aligned visual modalities, including RGB images, local depth maps, global semantic maps, and global depth maps.<n>Experiments on the JAAD and PIE benchmark datasets demonstrate that ADAPT consistently outperforms existing state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-07-14T03:05:09Z)
• MAIL++: Multi-Modal Bi-directional Agent Layer for Vision-Language Models [25.637441137217696]
  Multi-Modal Interactive Agent Layer embeds cross-modal coupling directly into intrinsic computation modules of vision-language models.<n>MAIL and MAIL++ consistently outperform state-of-the-art PEFT methods.
  arXiv  Detail & Related papers  (2026-05-25T06:35:33Z)
• RDNet: Region Proportion-Aware Dynamic Adaptive Salient Object Detection Network in Optical Remote Sensing Images [78.19052099452695]
  We propose the Region Proportion-Aware Dynamic Adaptive Salient Object Detection Network (RDNet)<n>RDNet replaces the CNN backbone with the SwinTransformer for global context modeling.<n>It achieves robustness against scale variations and accurate localization, delivering superior detection performance compared with state-of-the-art methods.
  arXiv  Detail & Related papers  (2026-03-12T17:34:29Z)
• Learnable Query Aggregation with KV Routing for Cross-view Geo-localisation [12.484512905649309]
  Cross-view geo-localisation (CVGL) aims to estimate the geographic location of a query image by matching it with images from a large-scale database.<n>To address these challenges, we propose a novel CVGL system that incorporates three key improvements.
  arXiv  Detail & Related papers  (2025-12-30T01:51:52Z)
• SC-Net: Robust Correspondence Learning via Spatial and Cross-Channel Context [19.20797236825297]
  Recent research has focused on using convolutional neural networks (CNNs) as the backbones in correspondence learning.<n>We propose a novel network named SC-Net, which effectively integrates bilateral context from both spatial and channel perspectives.<n>In experiments, SC-Net outperforms state-of-the-art methods in relative pose estimation and outlier removal tasks.
  arXiv  Detail & Related papers  (2025-12-29T13:56:10Z)
• Complementary and Contrastive Learning for Audio-Visual Segmentation [74.11434759171199]
  We present Complementary and Contrastive Transformer (CCFormer), a novel framework adept at processing both local and global information.<n>Our method sets new state-of-the-art benchmarks across the S4, MS3 and AVSS datasets.
  arXiv  Detail & Related papers  (2025-10-11T06:36:59Z)
• Multi-Level Aggregation and Recursive Alignment Architecture for Efficient Parallel Inference Segmentation Network [18.47001817385548]
  We propose a parallel inference network customized for semantic segmentation tasks. We employ a shallow backbone to ensure real-time speed, and propose three core components to compensate for the reduced model capacity to improve accuracy. Our framework shows a better balance between speed and accuracy than state-of-the-art real-time methods on Cityscapes and CamVid datasets.
  arXiv  Detail & Related papers  (2024-02-03T22:51:17Z)
• Adaptive Spot-Guided Transformer for Consistent Local Feature Matching [64.30749838423922]
  We propose Adaptive Spot-Guided Transformer (ASTR) for local feature matching. ASTR models the local consistency and scale variations in a unified coarse-to-fine architecture.
  arXiv  Detail & Related papers  (2023-03-29T12:28:01Z)
• PSNet: Parallel Symmetric Network for Video Salient Object Detection [85.94443548452729]
  We propose a VSOD network with up and down parallel symmetry, named PSNet. Two parallel branches with different dominant modalities are set to achieve complete video saliency decoding.
  arXiv  Detail & Related papers  (2022-10-12T04:11:48Z)
• Learning Target-aware Representation for Visual Tracking via Informative Interactions [49.552877881662475]
  We introduce a novel backbone architecture to improve target-perception ability of feature representation for tracking. The proposed GIM module and InBN mechanism are general and applicable to different backbone types including CNN and Transformer.
  arXiv  Detail & Related papers  (2022-01-07T16:22:27Z)
• RGB-D Salient Object Detection with Cross-Modality Modulation and Selection [126.4462739820643]
  We present an effective method to progressively integrate and refine the cross-modality complementarities for RGB-D salient object detection (SOD) The proposed network mainly solves two challenging issues: 1) how to effectively integrate the complementary information from RGB image and its corresponding depth map, and 2) how to adaptively select more saliency-related features.
  arXiv  Detail & Related papers  (2020-07-14T14:22:50Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.