Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers

CAAI Artificial Intelligence Research, Год журнала: 2023, Номер unknown, С. 9150015 - 9150015

Опубликована: Июнь 30, 2023

Most polyp segmentation methods use convolutional neural networks (CNNs) as their backbone, leading to two key issues when exchanging information between the encoder and decoder: (1) taking into account differences in contribution different-level features, (2) designing an effective mechanism for fusing these features. Unlike existing CNN-based methods, we adopt a transformer encoder, which learns more powerful robust representations. In addition, considering image acquisition influence elusive properties of polyps, introduce three standard modules, including cascaded fusion module (CFM), camouflage identification (CIM), similarity aggregation (SAM). Among these, CFM is used collect semantic location polyps from high-level features; CIM applied capture disguised low-level SAM extends pixel features area with position entire area, thereby effectively cross-level The proposed model, named Polyp-PVT, suppresses noises significantly improves expressive capabilities. Extensive experiments on five widely adopted datasets show that model various challenging situations (e.g., appearance changes, small objects, rotation) than representative methods. available at https://github.com/DengPingFan/Polyp-PVT.

Язык: Английский

---

Camouflaged Object Detection via Context-Aware Cross-Level Fusion

IEEE Transactions on Circuits and Systems for Video Technology, Год журнала: 2022, Номер 32(10), С. 6981 - 6993

Опубликована: Май 26, 2022

Camouflaged object detection (COD) aims to identify the objects that conceal themselves in natural scenes. Accurate COD suffers from a number of challenges associated with low boundary contrast and large variation appearances, e.g., size shape. To address these challenges, we propose novel Context-aware Cross-level Fusion Network ( $\text{C}^{2}\text{F}$ -Net), which fuses context-aware cross-level features for accurately identifying camouflaged objects. Specifically, compute informative attention coefficients multi-level our Attention-induced Module (ACFM), further integrates under guidance coefficients. We then Dual-branch Global Context (DGCM) refine fused feature representations by exploiting rich global context information. Multiple ACFMs DGCMs are integrated cascaded manner generating coarse prediction high-level features. The acts as an map low-level before passing them Camouflage Inference (CIM) generate final prediction. perform extensive experiments on three widely used benchmark datasets compare -Net state-of-the-art (SOTA) models. results show is effective model outperforms SOTA models remarkably. Further, evaluation polyp segmentation demonstrates promising potentials downstream applications. Our code publicly available at: https://github.com/Ben57882/C2FNet-TSCVT

Язык: Английский

---

Deep Gradient Learning for Efficient Camouflaged Object Detection

Deleted Journal, Год журнала: 2023, Номер 20(1), С. 92 - 108

Опубликована: Янв. 10, 2023

Abstract This paper introduces deep gradient network (DGNet), a novel framework that exploits object supervision for camouflaged detection (COD). It decouples the task into two connected branches, i.e., context and texture encoder. The essential connection is gradient-induced transition, representing soft grouping between features. Benefiting from simple but efficient framework, DGNet outperforms existing state-of-the-art COD models by large margin. Notably, our version, DGNet-S, runs in real-time (80 fps) achieves comparable results to cutting-edge model JCSOD-CVPR21 with only 6.82% parameters. application also show proposed performs well polyp segmentation, defect detection, transparent segmentation tasks. code will be made available at https://github.com/GewelsJI/DGNet .

Язык: Английский

---

Polyp-PVT: Polyp Segmentation with Pyramid Vision Transformers

arXiv (Cornell University), Год журнала: 2021, Номер unknown

Опубликована: Янв. 1, 2021

Most polyp segmentation methods use CNNs as their backbone, leading to two key issues when exchanging information between the encoder and decoder: 1) taking into account differences in contribution different-level features 2) designing an effective mechanism for fusing these features. Unlike existing CNN-based methods, we adopt a transformer encoder, which learns more powerful robust representations. In addition, considering image acquisition influence elusive properties of polyps, introduce three standard modules, including cascaded fusion module (CFM), camouflage identification (CIM), similarity aggregation (SAM). Among these, CFM is used collect semantic location polyps from high-level features; CIM applied capture disguised low-level features, SAM extends pixel area with position entire area, thereby effectively cross-level The proposed model, named Polyp-PVT, suppresses noises significantly improves expressive capabilities. Extensive experiments on five widely adopted datasets show that model various challenging situations (e.g., appearance changes, small objects, rotation) than representative methods. available at https://github.com/DengPingFan/Polyp-PVT.

Язык: Английский

---

Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications

Deleted Journal, Год журнала: 2024, Номер 21(4), С. 617 - 630

Опубликована: Апрель 12, 2024

Abstract Recently, Meta AI Research approaches a general, promptable segment anything model (SAM) pre-trained on an unprecedentedly large segmentation dataset (SA-1B). Without doubt, the emergence of SAM will yield significant benefits for wide array practical image applications. In this study, we conduct series intriguing investigations into performance across various applications, particularly in fields natural images, agriculture, manufacturing, remote sensing and healthcare. We analyze discuss limitations SAM, while also presenting outlook its future development tasks. By doing so, aim to give comprehensive understanding SAM’s This work is expected provide insights that facilitate research activities toward generic segmentation. Source code publicly available at https://github.com/LiuTingWed/SAM-Not-Perfect .

Язык: Английский

---

On the challenges and perspectives of foundation models for medical image analysis

Medical Image Analysis, Год журнала: 2023, Номер 91, С. 102996 - 102996

Опубликована: Окт. 12, 2023

Язык: Английский

---

2MGAS-Net: multi-level multi-scale gated attentional squeezed network for polyp segmentation

Signal Image and Video Processing, Год журнала: 2024, Номер 18(6-7), С. 5377 - 5386

Опубликована: Май 10, 2024

Язык: Английский

---

Assessing generalisability of deep learning-based polyp detection and segmentation methods through a computer vision challenge

Scientific Reports, Год журнала: 2024, Номер 14(1)

Опубликована: Янв. 23, 2024

Abstract Polyps are well-known cancer precursors identified by colonoscopy. However, variability in their size, appearance, and location makes the detection of polyps challenging. Moreover, colonoscopy surveillance removal highly operator-dependent procedures occur a complex organ topology. There exists high missed rate incomplete colonic polyps. To assist clinical reduce rates, automated methods for detecting segmenting using machine learning have been achieved past years. major drawback most these is ability to generalise out-of-sample unseen datasets from different centres, populations, modalities, acquisition systems. test this hypothesis rigorously, we, together with expert gastroenterologists, curated multi-centre multi-population dataset acquired six systems challenged computational teams develop robust segmentation crowd-sourcing Endoscopic computer vision challenge. This work put forward rigorous generalisability tests assesses usability devised deep dynamic actual procedures. We analyse results four top performing task five task. Our analyses demonstrate that top-ranking concentrated mainly on accuracy over real-time performance required applicability. further dissect provide an experiment-based reveals need improved tackle diversity present routine

Язык: Английский

---

A survey on deep learning for polyp segmentation: techniques, challenges and future trends

Visual Intelligence, Год журнала: 2025, Номер 3(1)

Опубликована: Янв. 3, 2025

Язык: Английский

---

ZoomNeXt: A Unified Collaborative Pyramid Network for Camouflaged Object Detection

IEEE Transactions on Pattern Analysis and Machine Intelligence, Год журнала: 2024, Номер 46(12), С. 9205 - 9220

Опубликована: Июнь 21, 2024

Recent camouflaged object detection (COD) attempts to segment objects visually blended into their surroundings, which is extremely complex and difficult in real-world scenarios.Apart from the high intrinsic similarity between background, are usually diverse scale, fuzzy appearance, even severely occluded.To this end, we propose an effective unified collaborative pyramid network that mimics human behavior when observing vague images videos, i.e., zooming out.Specifically, our approach employs strategy learn discriminative mixed-scale semantics by multi-head scale integration rich granularity perception units, designed fully explore imperceptible clues candidate background surroundings.The former's aggregation provides more visual patterns.The latter's routing mechanism can effectively propagate inter-frame differences spatiotemporal scenarios be adaptively deactivated output all-zero results for static representations.They provide a solid foundation realizing architecture dynamic COD.Moreover, considering uncertainty ambiguity derived indistinguishable textures, construct simple yet regularization, awareness loss, encourage predictions with higher confidence regions.Our highly task-friendly framework consistently outperforms existing state-of-the-art methods image video COD benchmarks.

Язык: Английский

---