Cross-Layer Attentive Feature Upsampling for Low-latency Semantic Segmentation

• URL: http://arxiv.org/abs/2601.01167v1
• Date: Sat, 03 Jan 2026 12:09:49 GMT
• Title: Cross-Layer Attentive Feature Upsampling for Low-latency Semantic Segmentation
• Authors: Tianheng Cheng, Xinggang Wang, Junchao Liao, Wenyu Liu,
• Abstract summary: We propose Guided Attentive Interpolation (GAI) to adaptively interpolate fine-grained high-resolution features with semantic features.<n>GAI determines both spatial and semantic relations of pixels from features of different resolutions and then leverages these relations to interpolate high-resolution features with rich semantics.<n>In experiments, the GAI-based semantic segmentation networks, i.e., GAIN, can achieve78.8 mIoU with 22.3 FPS on Cityscapes and 80.6 mIoU with 64.5 on CamVid using an NVIDIA 1080Ti GPU.
• Score: 52.01210390327581
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: Semantic segmentation is a fundamental problem in computer vision and it requires high-resolution feature maps for dense prediction. Current coordinate-guided low-resolution feature interpolation methods, e.g., bilinear interpolation, produce coarse high-resolution features which suffer from feature misalignment and insufficient context information. Moreover, enriching semantics to high-resolution features requires a high computation burden, so that it is challenging to meet the requirement of lowlatency inference. We propose a novel Guided Attentive Interpolation (GAI) method to adaptively interpolate fine-grained high-resolution features with semantic features to tackle these issues. Guided Attentive Interpolation determines both spatial and semantic relations of pixels from features of different resolutions and then leverages these relations to interpolate high-resolution features with rich semantics. GAI can be integrated with any deep convolutional network for efficient semantic segmentation. In experiments, the GAI-based semantic segmentation networks, i.e., GAIN, can achieve78.8 mIoU with 22.3 FPS on Cityscapes and 80.6 mIoU with 64.5 on CamVid using an NVIDIA 1080Ti GPU, which are the new state-of-the-art results of low-latency semantic segmentation. Code and models are available at: https://github.com/hustvl/simpleseg.

Related papers

• Scale-DiT: Ultra-High-Resolution Image Generation with Hierarchical Local Attention [50.391914489898774]
  Scale-DiT is a new diffusion framework that introduces hierarchical local attention with low-resolution global guidance.<n>A lightweight LoRA adaptation bridges global and local pathways during denoising, ensuring consistency across structure and detail.<n>Experiments demonstrate that Scale-DiT achieves more than $2times$ faster inference and lower memory usage.
  arXiv  Detail & Related papers  (2025-10-18T03:15:26Z)
• Graph-Based Uncertainty Modeling and Multimodal Fusion for Salient Object Detection [12.743278093269325]
  We propose a dynamic uncertainty propagation and multimodal collaborative reasoning network (DUP-MCRNet)<n>DUGC is designed to propagate uncertainty between layers through a sparse graph constructed based on spatial semantic distance.<n>MCF uses learnable modality gating weights to weightedly fuse the attention maps of RGB, depth, and edge features.
  arXiv  Detail & Related papers  (2025-08-28T04:31:48Z)
• Project-and-Fuse: Improving RGB-D Semantic Segmentation via Graph Convolution Networks [18.064378925844895]
  We propose to fuse features from two modalities in a late fusion style, during which the geometric feature injection is guided by texture feature prior.<n>At the 3D feature extraction stage, we argue that traditional CNNs are not efficient enough for depth maps.<n>At projection matrix generation stage, we find the existence of Biased-Assignment and Ambiguous-Locality issues in the original pipeline.
  arXiv  Detail & Related papers  (2025-01-31T02:24:13Z)
• Multi-Level Embedding and Alignment Network with Consistency and Invariance Learning for Cross-View Geo-Localization [2.733505168507872]
  Cross-View Geo-Localization (CVGL) involves determining the localization of drone images by retrieving the most similar GPS-tagged satellite images.<n>Existing methods often overlook the problem of increased computational and storage requirements when improving model performance.<n>We propose a lightweight enhanced alignment network, called the Multi-Level Embedding and Alignment Network (MEAN)
  arXiv  Detail & Related papers  (2024-12-19T13:10:38Z)
• Learning Implicit Feature Alignment Function for Semantic Segmentation [51.36809814890326]
  Implicit Feature Alignment function (IFA) is inspired by the rapidly expanding topic of implicit neural representations. We show that IFA implicitly aligns the feature maps at different levels and is capable of producing segmentation maps in arbitrary resolutions. Our method can be combined with improvement on various architectures, and it achieves state-of-the-art accuracy trade-off on common benchmarks.
  arXiv  Detail & Related papers  (2022-06-17T09:40:14Z)
• Boundary Corrected Multi-scale Fusion Network for Real-time Semantic Segmentation [15.879949436633021]
  Existing semantic segmentation methods rely on the high-resolution input to achieve high accuracy and do not meet the requirements of inference time. We propose a new method named Boundary Corrected Multi-scale Fusion Network, which uses the designed Low-resolution Multi-scale Fusion Module to extract semantic information. Our method achieves a state-of-the-art balance of accuracy and speed for the real-time semantic segmentation.
  arXiv  Detail & Related papers  (2022-03-01T13:31:01Z)
• A Holistically-Guided Decoder for Deep Representation Learning with Applications to Semantic Segmentation and Object Detection [74.88284082187462]
  One common strategy is to adopt dilated convolutions in the backbone networks to extract high-resolution feature maps. We propose one novel holistically-guided decoder which is introduced to obtain the high-resolution semantic-rich feature maps.
  arXiv  Detail & Related papers  (2020-12-18T10:51:49Z)
• Adaptive Linear Span Network for Object Skeleton Detection [56.78705071830965]
  We propose adaptive linear span network (AdaLSN) to automatically configure and integrate scale-aware features for object skeleton detection. AdaLSN substantiates its versatility by achieving significantly higher accuracy and latency trade-off. It also demonstrates general applicability to image-to-mask tasks such as edge detection and road extraction.
  arXiv  Detail & Related papers  (2020-11-08T12:51:14Z)
• Affinity Space Adaptation for Semantic Segmentation Across Domains [57.31113934195595]
  In this paper, we address the problem of unsupervised domain adaptation (UDA) in semantic segmentation. Motivated by the fact that source and target domain have invariant semantic structures, we propose to exploit such invariance across domains. We develop two affinity space adaptation strategies: affinity space cleaning and adversarial affinity space alignment.
  arXiv  Detail & Related papers  (2020-09-26T10:28:11Z)
• BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation [118.46210049742993]
  We propose an efficient and effective architecture with a good trade-off between speed and accuracy, termed Bilateral spatial Network (BiSeNet V2) For a 2,048x1, input, we achieve 72.6% Mean IoU on the Cityscapes test set with a speed of 156 FPS on one NVIDIA GeForce 1080 Ti card, which is significantly faster than existing methods, yet we achieve better segmentation accuracy.
  arXiv  Detail & Related papers  (2020-04-05T10:26:38Z)
• FarSee-Net: Real-Time Semantic Segmentation by Efficient Multi-scale Context Aggregation and Feature Space Super-resolution [14.226301825772174]
  We introduce a novel and efficient module called Cascaded Factorized Atrous Spatial Pyramid Pooling (CF-ASPP) It is a lightweight cascaded structure for Convolutional Neural Networks (CNNs) to efficiently leverage context information. We achieve 68.4% mIoU at 84 fps on the Cityscapes test set with a single Nivida Titan X (Maxwell) GPU card.
  arXiv  Detail & Related papers  (2020-03-09T03:53:57Z)

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.