DepthPolyp: Pseudo-Depth Guided Lightweight Segmentation for Real-Time Colonoscopy

• URL: http://arxiv.org/abs/2605.16519v1
• Date: Fri, 15 May 2026 18:14:32 GMT
• Title: DepthPolyp: Pseudo-Depth Guided Lightweight Segmentation for Real-Time Colonoscopy
• Abstract summary: We propose a lightweight and robust polyp segmentation framework based on pseudo-depth-guided multi-task learning and efficient feature modulation.<n>We demonstrate that DepthPolyp achieves strong cross-dataset generalization when trained on degraded data and evaluated on both clean and noisy target domains.<n>With only 3.57M parameters and 0.86 GMACs, the proposed method runs at over 180 FPS on mobile devices, making it well suited for real-time deployment in resource-constrained clinical environments.
• Score: 6.950410307869166
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate polyp segmentation in colonoscopy is essential for early colorectal cancer detection, yet real-world clinical environments pose persistent challenges such as motion blur, specular reflections, and illumination instability. Most existing methods are optimized on clean benchmark images and suffer noticeable performance degradation when deployed in authentic surgical scenarios. We propose DepthPolyp, a lightweight and robust segmentation framework based on pseudo-depth-guided multi-task learning and efficient feature modulation. The architecture combines hierarchical Ghost factorization for compact feature generation, Interleaved Shuffle Fusion for low-cost cross-scale interaction, and Dynamic Group Gating for adaptive group-wise feature weighting. Extensive experiments demonstrate that DepthPolyp achieves strong cross-dataset generalization when trained on degraded data and evaluated on both clean and noisy target domains, consistently outperforming lightweight baselines and remaining competitive with substantially larger models. In real surgical video evaluation on PolypGen, DepthPolyp achieves better segmentation performance than models up to $20\times$ larger while preserving real-time inference speed. With only 3.57M parameters and 0.86 GMACs, the proposed method runs at over 180 FPS on mobile devices, making it well suited for real-time deployment in resource-constrained clinical environments. Code and pretrained weights are available at: https://github.com/ReaganWu/DepthPolyp/

Related papers

• From SAM to DINOv2: Towards Distilling Foundation Models to Lightweight Baselines for Generalized Polyp Segmentation [9.452022523459886]
  Polyp-DiFoM is a framework that transfers the rich representations of foundation models into lightweight segmentation baselines.<n>It consistently outperforms baseline models significantly, as well as the state-of-the-art model, with nearly 9 times reduced overhead.
  arXiv  Detail & Related papers  (2025-12-10T04:25:01Z)
• CL-Polyp: A Contrastive Learning-Enhanced Network for Accurate Polyp Segmentation [2.000434989156371]
  We propose CL-Polyp, a contrastive learning-enhanced polyp segmentation network.<n>Our method uses contrastive learning to enhance the encoder's extraction of discriminative features.<n>It consistently surpasses state-of-the-art methods in clinical polyp segmentation.
  arXiv  Detail & Related papers  (2025-07-09T13:07:45Z)
• LGPS: A Lightweight GAN-Based Approach for Polyp Segmentation in Colonoscopy Images [40.19053450600923]
  We propose LGPS, a lightweight GAN-based framework for polyp segmentation.<n>LGPS incorporates three key innovations: (1) a MobileNetV2 backbone enhanced with modified residual blocks and Squeeze-and-Excitation modules for efficient feature extraction.<n>On the largest and challenging PolypGen test dataset, LGPS achieves a Dice of 0.7299 and an IoU of 0.7867, outperformed all SOTA works and demonstrating robust generalization.
  arXiv  Detail & Related papers  (2025-03-24T02:41:53Z)
• ASPS: Augmented Segment Anything Model for Polyp Segmentation [77.25557224490075]
  The Segment Anything Model (SAM) has introduced unprecedented potential for polyp segmentation. SAM's Transformer-based structure prioritizes global and low-frequency information. CFA integrates a trainable CNN encoder branch with a frozen ViT encoder, enabling the integration of domain-specific knowledge.
  arXiv  Detail & Related papers  (2024-06-30T14:55:32Z)
• Adaptation of Distinct Semantics for Uncertain Areas in Polyp Segmentation [11.646574658785362]
  This work presents a new novel architecture namely Adaptation of Distinct Semantics for Uncertain Areas in Polyp (ADSNet) ADSNet modifies misclassified details and recovers weak features having the ability to vanish and not be detected at the final stage. experimental results demonstrate the great correction and recovery ability leading to better segmentation performance compared to the other state of the art in the polyp image segmentation task.
  arXiv  Detail & Related papers  (2024-05-13T07:41:28Z)
• ECC-PolypDet: Enhanced CenterNet with Contrastive Learning for Automatic Polyp Detection [88.4359020192429]
  Existing methods either involve computationally expensive context aggregation or lack prior modeling of polyps, resulting in poor performance in challenging cases. In this paper, we propose the Enhanced CenterNet with Contrastive Learning (ECC-PolypDet), a two-stage training & end-to-end inference framework. Box-assisted Contrastive Learning (BCL) during training to minimize the intra-class difference and maximize the inter-class difference between foreground polyps and backgrounds, enabling our model to capture concealed polyps. In the fine-tuning stage, we introduce the IoU-guided Sample Re-weighting
  arXiv  Detail & Related papers  (2024-01-10T07:03:41Z)
• Recurrent Generic Contour-based Instance Segmentation with Progressive Learning [111.31166268300817]
  We propose a novel deep network architecture, i.e., PolySnake, for generic contour-based instance segmentation. Motivated by the classic Snake algorithm, the proposed PolySnake achieves superior and robust segmentation performance.
  arXiv  Detail & Related papers  (2023-01-21T05:34:29Z)
• Lesion-aware Dynamic Kernel for Polyp Segmentation [49.63274623103663]
  We propose a lesion-aware dynamic network (LDNet) for polyp segmentation. It is a traditional u-shape encoder-decoder structure incorporated with a dynamic kernel generation and updating scheme. This simple but effective scheme endows our model with powerful segmentation performance and generalization capability.
  arXiv  Detail & Related papers  (2023-01-12T09:53:57Z)
• Adaptive Context Selection for Polyp Segmentation [99.9959901908053]
  We propose an adaptive context selection based encoder-decoder framework which is composed of Local Context Attention (LCA) module, Global Context Module (GCM) and Adaptive Selection Module (ASM) LCA modules deliver local context features from encoder layers to decoder layers, enhancing the attention to the hard region which is determined by the prediction map of previous layer. GCM aims to further explore the global context features and send to the decoder layers. ASM is used for adaptive selection and aggregation of context features through channel-wise attention.
  arXiv  Detail & Related papers  (2023-01-12T04:06:44Z)
• PraNet: Parallel Reverse Attention Network for Polyp Segmentation [155.93344756264824]
  We propose a parallel reverse attention network (PraNet) for accurate polyp segmentation in colonoscopy images. We first aggregate the features in high-level layers using a parallel partial decoder (PPD) In addition, we mine the boundary cues using a reverse attention (RA) module, which is able to establish the relationship between areas and boundary cues.
  arXiv  Detail & Related papers  (2020-06-13T08:13:43Z)

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.