ATV-Net: Adaptive Triple-View Network with Dynamic Feature Fusion

• URL: http://arxiv.org/abs/2605.25803v1
• Date: Mon, 25 May 2026 12:52:01 GMT
• Title: ATV-Net: Adaptive Triple-View Network with Dynamic Feature Fusion
• Abstract summary: We propose ATV-Net, an Adaptive Triple-View Network that strengthens a ResNet-101 backbone using three simple but complementary receptive-field views.<n>Experiments on the Cityscapes validation set show that ATV-Net achieves 80.31% mIoU.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Recent semantic segmentation research has increasingly moved toward stronger context modeling, dense attention, and transformer-based architectures. Although these models achieve impressive performance, classical CNN-based segmentation pipelines remain attractive because of their simplicity, efficiency, and ease of implementation. This paper revisits a practical question: how far can a ResNet-based segmentation model be improved by only modifying the segmentation head? We propose ATV-Net, an Adaptive Triple-View Network that strengthens a ResNet-101 backbone using three simple but complementary receptive-field views. The micro view captures point-wise semantic responses, the local view models neighborhood structures and object boundaries, and the scout view provides enlarged contextual cues. Instead of fusing these views with fixed weights, ATV-Net introduces an Adaptive Decision Gate that dynamically selects receptive-field responses according to input scene characteristics. A compact global coordination layer is further applied to improve spatial and semantic consistency. Experiments on the Cityscapes validation set show that ATV-Net achieves 80.31\% mIoU. This result suggests that classical CNN-based segmentation is still far from obsolete: with simple receptive-field views and adaptive fusion, a ResNet-based pipeline can reach a competitive accuracy level without relying on transformer-style global attention or overly complex context modules.

Related papers

• Generalizing Geometry-Guided Mamba as a Plug-and-Play Context Module for CNN-based Semantic Segmentation [0.30786914102688595]
  This paper revisits Directional Geometric Mamba from DGM-Net and studies it as a plug-and-play context aggregation module.<n>The key idea is to inject geometric guidance into the selective scan process, allowing long-range feature propagation to be modulated by boundary and centripetal-flow cues.<n>Results on Cityscapes show consistent mIoU gains with only moderate extra GFLOPs at $1024times1024$ resolution.
  arXiv  Detail & Related papers  (2026-06-07T22:43:38Z)
• BMCR: Adaptive Backbone Module Composition via Reinforcement Learning for Remote Sensing Object Detection [4.303688954031907]
  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.
  arXiv  Detail & Related papers  (2026-06-04T02:04:15Z)
• RelayFormer: A Unified Local-Global Attention Framework for Scalable Image and Video Manipulation Localization [50.75654397516163]
  We propose RelayFormer, a unified framework that adapts to varying resolutions and modalities.<n> RelayFormer partitions inputs into fixed-size sub-images and introduces Global-Local Relay (GLR) tokens.<n>This enables efficient exchange of global cues, such as semantic or temporal consistency, while preserving fine-grained manipulation artifacts.
  arXiv  Detail & Related papers  (2025-08-13T03:35:28Z)
• ContextFormer: Redefining Efficiency in Semantic Segmentation [48.81126061219231]
  Convolutional methods, although capturing local dependencies well, struggle with long-range relationships.<n>Vision Transformers (ViTs) excel in global context capture but are hindered by high computational demands.<n>We propose ContextFormer, a hybrid framework leveraging the strengths of CNNs and ViTs in the bottleneck to balance efficiency, accuracy, and robustness for real-time semantic segmentation.
  arXiv  Detail & Related papers  (2025-01-31T16:11:04Z)
• EraW-Net: Enhance-Refine-Align W-Net for Scene-Associated Driver Attention Estimation [17.0226030258296]
  Associating driver attention with driving scene across two fields of views is a hard cross-domain perception problem. Previous methods typically focus on a single view or map attention to the scene via estimated gaze. We propose a novel method for end-to-end scene-associated driver attention estimation, called EraWNet.
  arXiv  Detail & Related papers  (2024-08-16T07:12:47Z)
• Vision Transformers: From Semantic Segmentation to Dense Prediction [139.15562023284187]
  We explore the global context learning potentials of vision transformers (ViTs) for dense visual prediction. Our motivation is that through learning global context at full receptive field layer by layer, ViTs may capture stronger long-range dependency information. We formulate a family of Hierarchical Local-Global (HLG) Transformers, characterized by local attention within windows and global-attention across windows in a pyramidal architecture.
  arXiv  Detail & Related papers  (2022-07-19T15:49:35Z)
• AA-RMVSNet: Adaptive Aggregation Recurrent Multi-view Stereo Network [8.127449025802436]
  We present a novel recurrent multi-view stereo network based on long short-term memory (LSTM) with adaptive aggregation, namely AA-RMVSNet. We firstly introduce an intra-view aggregation module to adaptively extract image features by using context-aware convolution and multi-scale aggregation. We propose an inter-view cost volume aggregation module for adaptive pixel-wise view aggregation, which is able to preserve better-matched pairs among all views.
  arXiv  Detail & Related papers  (2021-08-09T06:10:48Z)
• Container: Context Aggregation Network [83.12004501984043]
  Recent finding shows that a simple based solution without any traditional convolutional or Transformer components can produce effective visual representations. We present the model (CONText Ion NERtwok), a general-purpose building block for multi-head context aggregation. In contrast to Transformer-based methods that do not scale well to downstream tasks that rely on larger input image resolutions, our efficient network, named modellight, can be employed in object detection and instance segmentation networks.
  arXiv  Detail & Related papers  (2021-06-02T18:09:11Z)
• Conformer: Local Features Coupling Global Representations for Visual Recognition [72.9550481476101]
  We propose a hybrid network structure, termed Conformer, to take advantage of convolutional operations and self-attention mechanisms for enhanced representation learning. Experiments show that Conformer, under the comparable parameter complexity, outperforms the visual transformer (DeiT-B) by 2.3% on ImageNet.
  arXiv  Detail & Related papers  (2021-05-09T10:00:03Z)
• RPVNet: A Deep and Efficient Range-Point-Voxel Fusion Network for LiDAR Point Cloud Segmentation [28.494690309193068]
  We propose a novel range-point-voxel fusion network, namely RPVNet. In this network, we devise a deep fusion framework with multiple and mutual information interactions among these three views. By leveraging this efficient interaction and relatively lower voxel resolution, our method is also proved to be more efficient.
  arXiv  Detail & Related papers  (2021-03-24T04:24:12Z)
• Bottleneck Transformers for Visual Recognition [97.16013761605254]
  We present BoTNet, a powerful backbone architecture that incorporates self-attention for vision tasks. We present models that achieve a strong performance of 84.7% top-1 accuracy on the ImageNet benchmark. We hope our simple and effective approach will serve as a strong baseline for future research in self-attention models for vision.
  arXiv  Detail & Related papers  (2021-01-27T18:55:27Z)
• Visual Concept Reasoning Networks [93.99840807973546]
  A split-transform-merge strategy has been broadly used as an architectural constraint in convolutional neural networks for visual recognition tasks. We propose to exploit this strategy and combine it with our Visual Concept Reasoning Networks (VCRNet) to enable reasoning between high-level visual concepts. Our proposed model, VCRNet, consistently improves the performance by increasing the number of parameters by less than 1%.
  arXiv  Detail & Related papers  (2020-08-26T20:02:40Z)
• Gated Path Selection Network for Semantic Segmentation [72.44994579325822]
  We develop a novel network named Gated Path Selection Network (GPSNet), which aims to learn adaptive receptive fields. In GPSNet, we first design a two-dimensional multi-scale network - SuperNet, which densely incorporates features from growing receptive fields. To dynamically select desirable semantic context, a gate prediction module is further introduced.
  arXiv  Detail & Related papers  (2020-01-19T12:32:17Z)

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.