MambaPanoptic: A Vision Mamba-based Structured State Space Framework for Panoptic Segmentation

• URL: http://arxiv.org/abs/2605.12640v2
• Date: Tue, 19 May 2026 14:47:01 GMT
• Title: MambaPanoptic: A Vision Mamba-based Structured State Space Framework for Panoptic Segmentation
• Abstract summary: MambaPanoptic is a fully Mamba-based panoptic segmentation framework.<n>We introduce MambaFPN, a top-down feature pyramid that leverages Mamba blocks to generate globally coherent, multi-scale feature representations.<n>Experiments on the Cityscapes and panoptic segmentation benchmarks demonstrate that MambaPanoptic consistently outperforms PanopticDeepLab and PanopticFCN.
• Score: 44.88330743454271
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Panoptic segmentation requires the simultaneous recognition of countable thing instances and amorphous stuff regions, placing joint demands on long-range context modelling, multi-scale feature representation, and efficient dense prediction. Existing convolutional and transformer-based methods struggle to satisfy all three requirements concurrently: convolutional architectures are limited in their capacity to model long-range dependencies, while transformer-based methods incur quadratic computational cost that is prohibitive at high resolutions. In this paper, we propose MambaPanoptic, a fully Mamba-based panoptic segmentation framework that addresses these limitations through two principal contributions. First, we introduce MambaFPN, a top-down feature pyramid that leverages Mamba blocks to generate globally coherent, multi-scale feature representations with linear computational complexity. Second, we adopt a PanopticFCN-style kernel generator that produces unified thing and stuff kernels for proposal-free panoptic prediction, enhanced by a QuadMamba-based feature refinement module applied at multiple network stages. Experiments on the Cityscapes and COCO panoptic segmentation benchmarks demonstrate that MambaPanoptic consistently outperforms PanopticDeepLab and PanopticFCN under comparable model sizes, and matches or surpasses Mask2Former on Cityscapes in PQ and AP while requiring fewer parameters.

Related papers

• MambaADv2: Evolving Duality-enhanced State Space Model for Unsupervised Anomaly Detection [108.36437360254605]
  This paper proposes MambaADv2, a framework tailored for multi-class unsupervised anomaly detection.<n>MambaADv2 comprises a pre-trained encoder and a Mamba-inspired decoder, equipped with Duality-enhanced State Space (DSS) modules across multiple scales.<n>The structure of the Hybrid State Space (HSS) block is tailored by following the SSD-based Mamba lineage and incorporating Mamba3-style position-aware state-space modeling.
  arXiv  Detail & Related papers  (2026-06-22T10:14:06Z)
• MambaCount: Efficient Text-guided Open-vocabulary Object Counting with Spatial Sparse State Space Duality Block [8.77642380677869]
  Text-guided Open-vocabulary Object Counting (TOOC) aims to estimate the number of objects described by text prompts.<n>Mamba offers a promising alternative due to its linear complexity.<n>MambaCount is an efficient framework built on the Spatial Sparse State Space Duality (S4D) block.
  arXiv  Detail & Related papers  (2026-06-16T08:11:58Z)
• ForestMamba: Sparse Mamba with Geometry-guided Queries for 3D Forest Point Cloud Segmentation [9.432335594963956]
  ForestMamba is a structure-aware method that incorporates forest-specific priors into feature encoding, query generation, and query refinement.<n>Experiments across seven forest regions demonstrate that ForestMamba consistently outperforms existing baselines in both segmentation tasks.
  arXiv  Detail & Related papers  (2026-06-01T01:49:09Z)
• Towards Large Model Feature Coding [90.89390892560094]
  We present a benchmark and evaluation framework for large model feature coding (LaMoFC)<n>We first build the feature dataset LaMoFCBench, covering diverse task requirements across 4 categories and 16 scenarios.<n>We then specify representative split points according to practical application scenarios to extract intermediate features.
  arXiv  Detail & Related papers  (2026-05-20T13:39:48Z)
• Patch-MoE Mamba: A Patch-Ordered Mixture-of-Experts State Space Architecture for Medical Image Segmentation [11.412362266277675]
  We propose textitPatch-MoE Mamba, a patch-ordered mixture-of-experts state space architecture for medical image segmentation.<n>It introduces a hierarchical patch-ordered scanning mechanism that preserves local spatial neighborhoods while capturing multi-scale context.
  arXiv  Detail & Related papers  (2026-05-18T00:42:04Z)
• MambaBack: Bridging Local Features and Global Contexts in Whole Slide Image Analysis [8.432035888211212]
  Whole Slide Image (WSI) analysis is pivotal in computational pathology, enabling cancer diagnosis by integrating morphological and architectural cues.<n>Multiple Instance Learning (MIL) serves as the standard framework for WSI analysis.<n>MIL approaches face three critical challenges: (1) disruption of 2D spatial locality during 1D sequence flattening; (2) sub-optimal modeling of fine-grained local cellular structures; and (3) high memory peaks during inference.<n>We propose MambaBack, a novel hybrid architecture that harmonizes the strengths of Mamba and MambaOut.
  arXiv  Detail & Related papers  (2026-04-17T06:08:37Z)
• PanopMamba: Vision State Space Modeling for Nuclei Panoptic Segmentation [20.689908446030856]
  PanopMamba is a novel hybrid encoder-decoder architecture that integrates Mamba and Transformer.<n>To the best of our knowledge, this is the first Mamba-based approach for panoptic segmentation.<n>We introduce alternative evaluation metrics, including image-level Panoptic Quality ($i$PQ), boundary-weighted PQ ($w$PQ), and frequency-weighted PQ ($fw$PQ)
  arXiv  Detail & Related papers  (2026-01-23T10:33:15Z)
• PanFoMa: A Lightweight Foundation Model and Benchmark for Pan-Cancer [54.958921946378304]
  We introduce PanFoMa, a lightweight hybrid neural network that combines the strengths of Transformers and state-space models.<n>PanFoMa consists of a front-end local-context encoder with shared self-attention layers to capture complex, order-independent gene interactions.<n>We also construct a large-scale pan-cancer single-cell benchmark, PanFoMaBench, containing over 3.5 million high-quality cells.
  arXiv  Detail & Related papers  (2025-12-02T08:31:31Z)
• Topo-VM-UNetV2: Encoding Topology into Vision Mamba UNet for Polyp Segmentation [4.856498016044607]
  We propose Topo-VMUNetV2, which encodes topological features into the Mamba-based polyp segmentation model, VMUNetV2.<n>Our method consists two stages: VMUNetV2 is used to generate probability maps (PMs) for the training and test images, which are then used to compute topology attention maps.
  arXiv  Detail & Related papers  (2025-05-09T17:41:13Z)
• TransMamba: Fast Universal Architecture Adaption from Transformers to Mamba [66.80624029365448]
  We propose a cross-architecture knowledge transfer paradigm, TransMamba, that facilitates the reuse of Transformer pre-trained knowledge.<n>We propose a two-stage framework to accelerate the training of Mamba-based models, ensuring their effectiveness across both uni-modal and multi-modal tasks.
  arXiv  Detail & Related papers  (2025-02-21T01:22:01Z)
• Mamba-SEUNet: Mamba UNet for Monaural Speech Enhancement [54.427965535613886]
  Mamba, as a novel state-space model (SSM), has gained widespread application in natural language processing and computer vision.<n>In this work, we introduce Mamba-SEUNet, an innovative architecture that integrates Mamba with U-Net for SE tasks.
  arXiv  Detail & Related papers  (2024-12-21T13:43:51Z)
• PPMamba: A Pyramid Pooling Local Auxiliary SSM-Based Model for Remote Sensing Image Semantic Segmentation [1.5136939451642137]
  This paper proposes a novel network called Pyramid Pooling Mamba (PPMamba), which integrates CNN and Mamba for semantic segmentation tasks. PPMamba achieves competitive performance compared to state-of-the-art models.
  arXiv  Detail & Related papers  (2024-09-10T08:08:50Z)
• SIGMA: Selective Gated Mamba for Sequential Recommendation [56.85338055215429]
  Mamba, a recent advancement, has exhibited exceptional performance in time series prediction.<n>We introduce a new framework named Selective Gated Mamba ( SIGMA) for Sequential Recommendation.<n>Our results indicate that SIGMA outperforms current models on five real-world datasets.
  arXiv  Detail & Related papers  (2024-08-21T09:12:59Z)
• SPMamba: State-space model is all you need in speech separation [20.168153319805665]
  CNN-based speech separation models face local receptive field limitations and cannot effectively capture long time dependencies. We introduce an innovative speech separation method called SPMamba. This model builds upon the robust TF-GridNet architecture, replacing its traditional BLSTM modules with bidirectional Mamba modules.
  arXiv  Detail & Related papers  (2024-04-02T16:04:31Z)

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.