285+ 即插即用深度学习模块

通用模块

注意力机制 (54个)

Circulant Attention Learners

🎯 Vision Transformer/通用CV 🏛 AAAI 2026

Structured Awareness: Directional, Frequency-Spatial, and Structural Attention

🎯 医学图像分割 🏛 AAAI 2026

DCMM-Transformer: Degree-Corrected Mixed-Membership Attention

🎯 医学影像 🏛 AAAI 2026

ABDUCTIVEMLLM: Boosting Visual Abductive Reasoning

🎯 多模态推理 🏛 AAAI 2026

MODA: Multispectral Object Detection in Aerial Images

🎯 多光谱目标检测 🏛 AAAI 2026

UCMNet: Uncertainty-Aware Context Memory Network

🎯 屏下摄像头图像恢复 (水平-垂直双注意力) 🏛 CVPR 2026

VideoFusion: Spatio-Temporal Collaborative Network

🎯 多模态视频融合 (差分增强注意力) 🏛 CVPR 2026

Flickerformer: Periodicity and Directionality for Burst Flicker Removal

🎯 图像去闪烁 (小波注意力) 🏛 CVPR 2026

HVI: A New Color Space for Low-light Image Enhancement

🎯 低光增强 (跨注意力) 🏛 CVPR 2025

U-RWKV: Direction-Adaptive Lightweight Medical Segmentation

🎯 医学图像分割 (挤压激励注意力) 🏛 MICCAI 2025

MANO: Multipole Attention Mechanism

🎯 CV/物理 (多极注意力) 🏛 ICCV 2025

Probability-Guided Edge Enhancement Network

🎯 遥感语义分割 (卷积自注意力) 🏛 TGRS 2025

CTOD: Cross-Attentive Task-Alignment

🎯 单阶段目标检测 (任务交叉注意力) 🏛 TMM 2024

SADT: Scale-Adaptive Deformable Transformer

🎯 图像恢复 (尺度自适应可变形注意力) 🏛 CVPR 2025

Wavelet and Adaptive Coordinate Attention

🎯 图像去噪 (小波注意力) 🏛 TIM 2024

FreqSal: Fourier-embedded Network for RGB-T SOD

🎯 RGB-T显著性检测 (傅里叶残差注意力) 🏛 AAAI 2025

VolFormer: Cube Interaction for HSI Restoration

🎯 高光谱恢复 (三维立体注意力) 🏛 CVPR 2025

TBSN: Transformer Blind-Spot Network

🎯 自监督图像去噪 (通道注意力) 🏛 AAAI 2025

MCA: Multi-dimensional Collaborative Attention

🎯 通用CV (多维协作注意力) 🏛 EAAI 2023

MCANet: Multi-Scale Cross-Axis Attention

🎯 医学图像分割 (CV全领域通用) 🏛 arXiv 2023

RGT: Recursive Generalization Transformer (RG_SA)

🎯 图像超分 (递归泛化自注意力) 🏛 arXiv 2023

Energy-Based Cross Attention

🎯 文本到图像扩散模型 🏛 arXiv 2023

HiLo Attention

🎯 CV 2D (结合高频低频注意力) 🏛 NeurIPS 2022

PPA: Parallelized Patch-Aware Attention (HCF-Net)

🎯 红外小目标检测 (CV 2D通用) 🏛 arXiv 2024

AGCA: Adaptive Graph Channel Attention

🎯 CV 2D/图卷积 (钢表面缺陷检测) 🏛 TIM 2023

RGA: Relation-Aware Global Attention

🎯 行人重识别 (关系感知全局注意力) 🏛 CVPR 2020

EGA: Edge-Guided Attention (EGCIFFNet)

🎯 边缘检测/CV 2D图像任务 🏛 TIM 2024

Agent Attention: Softmax + Linear Attention

🎯 CV 2D通用 (全新注意力范式) 🏛 ECCV 2024

SENet: Squeeze-and-Excitation Networks (3D版本)

🎯 3D CV (通道注意力) 🏛 CVPR 2018

scSE: Spatial and Channel Squeeze & Excitation

🎯 图像分割 (空间通道注意力) 🏛 MICCAI 2018

GCT: Gated Channel Transformation

🎯 CV (改进通道注意力) 🏛 CVPR 2020

DICAM: Underwater Image Enhancement Attention

🎯 水下图像增强 🏛 Science TM 2022

HAAM: Hybrid Adaptive Attention Module (AAU-net)

🎯 医学图像分割 🏛 arXiv 2022

HWMNet: Half Wavelet Attention on M-Net+

🎯 低光图像增强 🏛 arXiv 2022

UBRFC-Net: Adaptive Fine-Grained Channel Attention

🎯 图像去雾 (改进SE通道注意力) 🏛 Neural Networks 2024

SCSA: Spatial and Channel Synergistic Attention

🎯 通用CV (空间通道协同注意力) 🏛 arXiv 2024

ENLTB: Efficient Non-Local Attention (Perspective+ Unet)

🎯 医学图像分割 🏛 MICCAI 2024

MLLA: Linear Attention from Mamba Perspective

🎯 CV通用 (继承Mamba优势的线性注意力) 🏛 arXiv 2024

LDConv: Linear Deformable Convolution Attention

🎯 CV通用 (线性可变形卷积注意力) 🏛 arXiv 2023

Haar Wavelet High-Low Frequency Attention

🎯 裂缝检测 (基于Haar小波注意力) 🏛 ESWA 2024

LGAG: Large Kernel Grouped Attention Gate

🎯 医学图像分割 (大核分组注意力门控) 🏛 CVPR 2024

CGLU: Convolutional Gated Linear Unit (TransNeXt)

🎯 CV/NLP通用 (卷积门控通道注意力) 🏛 CVPR 2024

MLKA: Multi-scale Large Kernel Attention (MAN)

🎯 CV 2D通用 (多尺度大核注意力) 🏛 CVPR 2024

DAT: Deformable Attention Transformer

🎯 时间序列预测 (可变形注意力) 🏛 CVPR 2022

FECAM: Frequency Enhanced Channel Attention

🎯 时间序列预测 (频率增强通道注意力) 🏛 arXiv 2022

DSANet: Dual Self-Attention Network

🎯 时间序列预测 (去稳态注意力) 🏛 CIKM 2019

Local Flow Attention

🎯 交通流预测 (局部流注意力) 🏛 Neural Networks 2023

DCT-Former: Self-Attention with DCT

🎯 时间序列/NLP (离散余弦变换注意力) 🏛 arXiv 2022

AGF: Attention Gate Fusion (MotionAGFormer)

🎯 3D人体姿态估计 (AGF注意力) 🏛 WACV 2024

RMT: Retentive Networks Meet Vision Transformers

🎯 CV通用 (保留网络注意力) 🏛 CVPR 2024

CATANet: Content-Aware Token Aggregation

🎯 轻量级图像超分 🏛 CVPR 2025

FSTA-SNN: Frequency-based Spatial-Temporal Attention

🎯 脉冲神经网络 (频域时空注意力) 🏛 AAAI 2025

HSPAN: High-Similarity-Pass Attention

🎯 图像超分 🏛 TIP 2024

PMFSNet: Polarized Multi-scale Feature Self-attention

🎯 轻量级医学图像分割 🏛 arXiv 2024

卷积模块 (24个)

Partial Channel Network

🎯 轻量级CNN (部分通道卷积) 🏛 AAAI 2026

Strip R-CNN: Large Strip Convolution

🎯 遥感目标检测 (大条带卷积) 🏛 AAAI 2026

Remote Sensing Forestry Similarity Convolution

🎯 遥感林业分类 🏛 WACV 2026

TM-BSN: Triangular-Masked Blind-Spot Network

🎯 自监督图像去噪 🏛 CVPR 2026

SCT-Net: CNN-Transformer Pooling Attention Fusion

🎯 高光谱分类 (2D+3D并行卷积) 🏛 DSP 2025

MobileIE: Lightweight ConvNet for Mobile IE

🎯 移动端实时图像增强 🏛 ICCV 2025

ARConv: Adaptive Rectangular Convolution

🎯 遥感全色锐化 (自适应卷积) 🏛 CVPR 2025

ConverseNet: Reverse Convolution

🎯 图像恢复 (反卷积算子) 🏛 ICCV 2025

Pinwheel-shaped Convolution

🎯 红外小目标检测 🏛 AAAI 2025

DEA-Net: Detail-Enhanced Convolution + Content-Guided Attention

🎯 图像去雾/CV 2D通用 🏛 TIP 2024

CTR-GC: Channel-wise Topology Refinement Graph Conv

🎯 骨架动作识别 (通道拓扑细化图卷积) 🏛 ICCV 2021

WTConv: Wavelet Convolutions for Large Receptive Fields

🎯 CV 2D通用 (小波变换卷积) 🏛 ECCV 2024

TVConv: Translation Variant Convolution

🎯 医学分割/人脸识别 (平移变体卷积) 🏛 CVPR 2022

Dynamic Convolution: Attention over Kernels

🎯 CV通用 (1D/2D/3D动态卷积) 🏛 CVPR 2020

PyConv: Pyramidal Convolution

🎯 CV通用 (金字塔卷积) 🏛 arXiv 2020

Multi-Dilation Rate Channel Convolution

🎯 目标检测 (多膨胀率通道卷积) 🏛 arXiv 2024

CondConv: Conditionally Parameterized Convolutions

🎯 CV通用 (经典动态卷积) 🏛 NeurIPS 2019

DO-Conv: Depthwise Over-parameterized Conv

🎯 CV通用 (替代传统卷积) 🏛 arXiv 2020

FasterNet: Partial Convolution (PConv)

🎯 轻量级CV 🏛 CVPR 2023

Large Kernel Convolution Downsampling

🎯 CV通用 (大核卷积下采样) 🏛 arXiv 2022

LDConv: Linear Deformable Convolution

🎯 CV通用 (线性可变形卷积) 🏛 SCI 2024

AKConv: Arbitrary Kernel Convolution

🎯 CV通用 (任意采样形状卷积) 🏛 arXiv 2023

BHViT: Binarized Hybrid Vision Transformer

🎯 轻量级CV (二值化混合ViT) 🏛 CVPR 2025

FADformer: Frequency-Domain Image Deraining

🎯 图像去雨 (频域卷积) 🏛 ECCV 2024

频域 (1个)

SFM: Spatial Frequency Modulation

🎯 语义分割 (空间频率调制) 🏛 TPAMI 2026

特征提取 (32个)

Cross-Modality Feature Adaptive Interaction

🎯 RGB-红外航空目标检测 (跨模态特征自适应) 🏛 TGRS 2026

Mesoscopic Insights: Multi-scale & Hybrid Architecture

🎯 图像篡改定位 🏛 AAAI 2025

Flora-NET: Dual Coordinate Attention + Adaptive Kernel

🎯 药用花卉识别 🏛 Elsevier 2025

Real-World Remote Sensing Image Dehazing

🎯 遥感图像去雾 🏛 TGRS 2025

I2U-Net: Dual-Path U-Net with MFII

🎯 医学分割 (双分支信息交互特征提取) 🏛 MedIA 2024

MixDehazeNet: Mix Structure Block

🎯 图像去雾 🏛 arXiv 2023

SCConv: Spatial and Channel Reconstruction Convolution

🎯 特征冗余压缩 (CV通用) 🏛 CVPR 2023

SCSegamba: Lightweight Structure-Aware Vision Mamba

🎯 裂缝分割 (SAVSS模块+MFS头) 🏛 CVPR 2025

DCMPNet: Depth Information Assisted Collaborative Network

🎯 单图像去雾 🏛 CVPR 2024

OTTER: Text-Aware Visual Feature Extraction VLA

🎯 机器人操作 (文本感知视觉特征) 🏛 ICML 2025

LogicAD: VLM-based Text Feature Extraction

🎯 异常检测 (可解释VLM特征) 🏛 AAAI 2025

DDM Deconstruction for Self-Supervised Learning (l-DAE)

🎯 自监督特征学习 (潜在去噪自编码器) 🏛 ICLR 2025

Diffusion Models for Sketch-Photo Matching

🎯 零样本草图检索 (扩散模型特征提取) 🏛 CVPR 2024

LFG-Diffusion: Latent Feature-Guided Diffusion

🎯 阴影去除 (潜在特征引导扩散) 🏛 WACV 2024

Weak-Mamba-UNet: CNN+ViT+Mamba Hybrid

🎯 涂鸦监督医学分割 (三架构协同) 🏛 arXiv 2024

BEFUnet: Hybrid CNN-Transformer Architecture

🎯 医学图像分割 🏛 arXiv 2024

CCT-LSTM: Compact CNN Transformer + LSTM

🎯 远程压力估计 (多模态特征) 🏛 WACV 2024

T-FREX: Transformer-based Feature Extraction

🎯 移动应用评论特征提取 (NER) 🏛 arXiv 2024

TR-DETR: Task-Reciprocal Transformer

🎯 视频时刻检索+高光检测 (多模态对齐) 🏛 AAAI 2024

MambaVision: Hybrid Mamba-Transformer Backbone

🎯 视觉主干网络 (Mamba+Transformer混合) 🏛 arXiv 2024

SaTQA: Transformer-based NR-IQA

🎯 无参考图像质量评估 (监督对比学习) 🏛 AAAI 2024

Tri-VAE: Triplet Variational Autoencoder

🎯 脑肿瘤MRI异常检测 (无监督) 🏛 CVPRW 2024

TSLANet: Time Series Lightweight Adaptive Network

🎯 时序特征提取 (自适应频谱块+交互卷积) 🏛 arXiv 2024

GraphKAN: Graph Kolmogorov Arnold Networks

🎯 图特征提取 (KAN增强) 🏛 arXiv 2024

MFDS-DETR: Multi-Level Feature Fusion + Deformable-DETR

🎯 白细胞检测 (多尺度特征融合) 🏛 arXiv 2024

Efficient LoFTR: Semi-Dense Feature Matching

🎯 图像匹配 (聚合注意力+两阶段相关) 🏛 CVPR 2024

XFeat: Accelerated Lightweight Image Matching

🎯 轻量级图像匹配 (资源受限设备) 🏛 CVPR 2024

FourierKAN-GCF: Fourier KAN for Graph CF

🎯 图协同过滤推荐 (傅里叶KAN特征变换) 🏛 arXiv 2024

LISN: Lightweight Information Split Network

🎯 红外图像超分 🏛 arXiv 2024

WaveNet-SF: Wavelet Spatial-Frequency Network

🎯 视网膜疾病检测 (小波变换空频域) 🏛 arXiv 2025

MLP-KAN: Deep Representation + Function Learning

🎯 通用深度学习 (MLP+KAN统一) 🏛 arXiv 2024

WaveletMamba (W-Mamba): Wavelet + SSM Fusion

🎯 红外-可见光图像融合 🏛 arXiv 2025

特征融合 (14个)

FAAFusion: Fourier Angle Alignment

🎯 遥感旋转目标检测 (傅里叶频域特征融合) 🏛 CVPR 2026

LFSB: Differential Dual-Stream Attention (ReflexSplit)

🎯 反射分离 (差分双流注意力融合) 🏛 CVPR 2026

TransMixer: CNN+Transformer+Mamba Architecture

🎯 裂缝分割 (三架构协同特征融合) 🏛 CVPR 2026

D2T: Dual-Domain Feature Fusion (WPFormer)

🎯 缺陷检测/小目标检测 (双域特征融合) 🏛 CVPR 2025

Dynamic Feature Fusion for Emotional Mimicry

🎯 情感模仿强度估计 (跨模态动态融合) 🏛 CVPRW 2025

ConDSeg: Contrast-Driven Feature Enhancement

🎯 医学图像分割 (对比驱动特征增强融合) 🏛 AAAI 2025

Haar Wavelet High-Low Frequency Attention Fusion

🎯 裂缝分割 (Haar小波高低频融合) 🏛 ESWA 2024

DFF: Dynamic Feature Fusion

🎯 语义边缘检测 (动态特征融合/2D+3D) 🏛 arXiv 2019

DASI: Hierarchical Context Fusion (HCF-Net)

🎯 红外小目标检测 (特征融合) 🏛 arXiv 2024

SDM: Feature Fusion for Segmentation (PnPNet)

🎯 3D医学分割 (特征融合/2D+3D) 🏛 arXiv 2023

TIF: Transformer Interaction Fusion (DS-TransUNet)

🎯 医学分割 (跳跃连接特征融合) 🏛 arXiv 2021

SFFusion: Semantic-Aware Feature Fusion

🎯 红外-可见光融合 (语义感知/2D+3D) 🏛 Information Fusion 2022

CGAFusion: Content-Guided Attention Fusion (DEA-Net)

🎯 图像去雾 (低级+高级特征融合) 🏛 TIP 2024

GLSA: Global-Local Spatial Feature Fusion (DuAT)

🎯 医学分割/CV通用 (全局-局部空间融合) 🏛 PRCV 2023

下采样 (8个)

ASCNet: Asymmetric Sampling Correction

🎯 红外图像去条纹 (非对称采样校正) 🏛 TIM 2025

Down-Sampling Rollouts in LLM RL

🎯 LLM强化学习 (下采样优化) 🏛 arXiv 2025

DABI: Downsampling in Bilateral Control Imitation

🎯 模仿学习数据增强 🏛 arXiv 2024

Group Downsampling with Equivariant Anti-Aliasing

🎯 CV通用 (等变抗混叠群下采样) 🏛 ICLR 2025

DS-Pnet: Downsampling Positioning

🎯 FM定位 (下采样定位) 🏛 arXiv 2025

ADAPTOR: Adaptive Token Reduction

🎯 视频扩散Transformer (自适应Token下采样) 🏛 CVPRW 2025

Dynamic U-Net: Adaptive Feature Calibration

🎯 腹部多器官分割 (自适应下采样) 🏛 arXiv 2024

UAV-DETR: End-to-End Object Detection

🎯 无人机图像检测 (高效下采样) 🏛 arXiv 2025

归一化 (10个)

BCN: Batch Channel Normalization

🎯 图像分类 (批通道归一化) 🏛 arXiv 2023

Lipschitz Normalization

🎯 GAT/Graph Transformer (Lipschitz归一化) 🏛 ICML 2021

CrossNorm + SelfNorm

🎯 OOD鲁棒性 (两种归一化方式) 🏛 ICCV 2021

ContraNorm: Contrastive Normalization

🎯 GNN/Transformer (对比归一化层) 🏛 ICLR 2023

DyT: Transformers without Normalization

🎯 Transformer (替代归一化) 🏛 CVPR 2025

DiMR: Multi-Resolution Diffusion + Time-Dependent LN

🎯 图像生成 (时间依赖层归一化) 🏛 NeurIPS 2024

TRIBE: Tri-net Self-Training with Balanced Norm

🎯 测试时自适应 (平衡归一化) 🏛 AAAI 2024

MABN: Domain-Aware Batch Normalization

🎯 测试时域自适应 🏛 AAAI 2024

SN-DCR: Spectral Normalization + Dual Contrastive

🎯 图像到图像翻译 🏛 arXiv 2023

Hyperspherical Normalization for DRL

🎯 深度强化学习 (超球面归一化) 🏛 arXiv 2025

多尺度融合 (12个)

DCCS-Det: Directional Cross-Scale Detector

🎯 红外小目标检测 (方向上下文跨尺度) 🏛 TGRS 2026

DTP: Dual-Path Frequency Structural Decoupling

🎯 低光超分 (频域结构解耦双路径) 🏛 ICME 2026

FBRT-YOLO: Real-Time Aerial Detection

🎯 实时航空图像检测 🏛 AAAI 2025

Lightweight Multiscale Feature Fusion

🎯 航空小目标检测 (轻量级多尺度融合) 🏛 TGRS 2025

GLVMamba: Global-Local Visual State-Space Model

🎯 遥感分割 (全局-局部多尺度融合) 🏛 TGRS 2025

HISRCNet: SR + Classification for Histopathology

🎯 乳腺癌病理图像超分+分类 🏛 MICCAI 2023

CEDNET: Cascade Encoder-Decoder Network

🎯 密集预测 (级联编码-解码) 🏛 ICLR 2023

AMD: Adaptive Multi-Scale Decomposition

🎯 时间序列预测 (自适应多尺度分解) 🏛 AAAI 2025

MF-Mamba: Multi-scale Mamba Fusion

🎯 遥感语义分割 (多尺度Mamba融合) 🏛 TGRS 2025

MDFM: Multi-Decision Fusing Model

🎯 遥感变化检测 (多尺度差异融合) 🏛 TGRS 2024

DFF: Dynamic Feature Fusion (D-Net/DLK)

🎯 3D医学分割 (多尺度动态特征融合) 🏛 arXiv 2024

CCFF: Cross-Scale Feature Fusion (RT-DETR)

🎯 实时目标检测 (跨尺度特征融合) 🏛 CVPR 2024

上采样 (1个)

DySample: Learning to Upsample by Learning to Sample

🎯 CV 2D通用 (动态上采样) 🏛 ICCV 2023

轻量化 (16个)

MobileNetV4: Universal Models for Mobile

🎯 移动端通用 (UIB块) 🏛 arXiv 2024

FMViT: Multiple-Frequency Mixing ViT

🎯 轻量级视觉主干 (高低频混合) 🏛 arXiv 2023

Rethinking Attention: Shallow MLP Alternative

🎯 轻量级Transformer (MLP替换注意力) 🏛 AAAI 2024

Frequency-Enhanced Feature Distillation

🎯 频域增强特征蒸馏轻量化 🏛 ACM MM 2022

MobileDenseNet: Lightweight Object Detection

🎯 移动端目标检测 🏛 arXiv 2022

Skip-Attention: Paying Less Attention

🎯 ViT轻量化 (降低计算量) 🏛 arXiv 2023

SHViT: Single-Head ViT

🎯 轻量级ViT (碾压MobileNet/ShuffleNet) 🏛 arXiv 2024

Lightweight Stacked Hourglass Network

🎯 视觉感知 (轻量化沙漏网络) 🏛 arXiv 2023

EfficientViT: Cascaded Group Attention

🎯 ViT高效部署 (级联分组注意力) 🏛 CVPR 2023

Focus-DETR: Less is More

🎯 轻量化DETR (华为诺亚) 🏛 ICCV 2023

FalconNet: Lightweight ConvNet Factorization

🎯 轻量级Backbone (汇集所有轻量化优点) 🏛 arXiv 2023

MobileViT: Lightweight Mobile ViT

🎯 移动端ViT (CNN+Self-Attention融合) 🏛 ICLR 2022

EdgeNeXt: CNN-Transformer for Mobile Vision

🎯 移动视觉 (Channel Attention增强CNN) 🏛 ECCVW 2022

EfficientFormer: ViTs at MobileNet Speed

🎯 移动端Transformer (MobileNet速度) 🏛 NeurIPS 2022

TinyViT: Fast Pretraining Distillation

🎯 小型ViT (蒸馏预训练) 🏛 ECCV 2022

MobileOne: 1ms Mobile Backbone

🎯 手机端1ms级主干网 🏛 CVPR 2023

损失函数 (1个)

Artifact Regularization + Walsh-Hadamard Transform

🎯 低光图像增强 (新损失函数) 🏛 ACM 2025

Backbone (2个)

StarNet: Rewrite the Stars

🎯 通用CV Backbone (元素相乘>相加) 🏛 CVPR 2024

GhostNetV2: GhostModule V1&V2

🎯 CV通用 (替代传统卷积的Ghost模块) 🏛 NeurIPS 2022

前沿技术专项

KAN (2个)

KAN: Kolmogorov-Arnold Networks

🎯 通用深度学习 (KAN缝合操作指南) 🏛 arXiv 2024

SCKansformer: KAN + SCConv Backbone

🎯 骨髓细胞细粒度分类 (KAN+SCConv) 🏛 arXiv 2024

Mamba (18个)

RFGM: Beyond Illumination for Extreme Dark Restoration

🎯 极暗图像恢复 🏛 AAAI 2026

C2SSM: Cluster-Centric Scan for UHD Restoration

🎯 超高清图像恢复 (聚类中心扫描) 🏛 CVPR 2026

MaIR: Locality- and Continuity-Preserving Mamba

🎯 图像恢复 (局部连续保持Mamba) 🏛 CVPR 2025

SCSegamba: Lightweight Structure-Aware Mamba

🎯 裂缝分割 (结构感知轻量Mamba) 🏛 CVPR 2025

MobileMamba: Lightweight Multi-Receptive Mamba

🎯 轻量级多感受野视觉Mamba 🏛 CVPR 2025

MambaHSI: Spatial-Spectral Mamba for HSI

🎯 高光谱图像分类 (空间-光谱Mamba) 🏛 TGRS 2024

EfficientViM: Efficient Vision Mamba

🎯 CV通用 (Hidden State Mixer SSD) 🏛 CVPR 2025

ConvSSM: Convolutional State Space Models

🎯 CV 2D通用 (卷积状态空间模型) 🏛 NeurIPS 2023

nnMamba: 3D Biomedical SSM

🎯 3D医学图像分割/分类/关键点检测 🏛 arXiv 2024

TimeMachine: 4 Mambas for Time Series

🎯 时间序列长期预测 🏛 arXiv 2024

MambaIR: Mamba for Image Restoration (RSSG)

🎯 图像恢复 (通道注意力+局部增强Mamba) 🏛 ECCV 2024

RSCaMa: Remote Sensing Change Captioning Mamba

🎯 遥感变化检测/视频理解 (联合时空Mamba) 🏛 arXiv 2024

Jamba: Hybrid Transformer-Mamba LM

🎯 CV+NLP通用 (混合Transformer-Mamba) 🏛 arXiv 2024

VMamba: Visual State Space Model (PVMamba)

🎯 医学分割/CV通用 (并行化视觉Mamba) 🏛 NeurIPS 2024

CM-UNet: Hybrid CNN-Mamba UNet (CSMamba)

🎯 遥感语义分割/CV通用 (CSMamba解码器) 🏛 arXiv 2024

SegMamba: 3D Medical Image Mamba

🎯 3D医学分割 (2D+3D Mamba卷积) 🏛 arXiv 2024

WalMaFa: Wavelet Mamba with Fourier Adjustment

🎯 低光图像增强 (小波Mamba+傅里叶调整) 🏛 arXiv 2024

MambaOut: Do We Really Need Mamba for Vision?

🎯 CV (Mamba必要性探索) 🏛 CVPR 2025

扩散模型 (2个)

FreeU: Free Lunch in Diffusion U-Net

🎯 扩散模型改进U-Net (无需训练) 🏛 CVPR 2024

KSA-Edit: All-in-One Slider for Diffusion

🎯 扩散模型图像属性编辑 (轻量级模块) 🏛 CVPR 2026

细分任务场景

多模态 (1个)

STC: Multispectral Sensors Color Correction

🎯 手机相机颜色校正 (多光谱传感器) 🏛 CVPR 2026

时间序列 (15个)

FusionRegister: IVIF Registration

🎯 红外-可见光图像融合配准 🏛 CVPR 2026

Hierarchical Token Compression for Streaming VLLM

🎯 流式视频大语言模型加速 🏛 CVPR 2026

CPUBone: Efficient Vision Backbone

🎯 低并行能力设备视觉主干 🏛 CVPR 2026

TimeBase: Minimalist Long-term Forecasting

🎯 高效长期时间序列预测 🏛 ICML 2025

DyT: Transformers without Normalization

🎯 Transformer时序建模 (动态Tanh替代) 🏛 CVPR 2025

TERSE: Temporal Restoration + Spatial Rewiring

🎯 无源多变量时序域自适应 🏛 KDD 2025

MLOW: Low-Rank Frequency Decomposition

🎯 时间序列预测 (多效应频率分解) 🏛 arXiv 2025

FreqEvo: Multi-Level Frequency Feature Extraction

🎯 时间序列预测 (多级频域特征) 🏛 TKDE 2025

CORA: Covariate-Aware Adaptation

🎯 时序基础模型协变量自适应 🏛 ICLR 2026

FITS: Time Series with 10k Parameters

🎯 轻量级时序预测 (频域角度) 🏛 ICLR 2024 Spotlight

SST: Multi-Scale Hybrid Mamba-Transformer

🎯 长短期时序预测 (Mamba-Transformer混合) 🏛 CIKM 2025

CrossLinear: Cross-Correlation Embedding

🎯 外生变量时序预测 (即插即用) 🏛 arXiv 2025

TSLANet: Rethinking Transformers for TS

🎯 时序表示学习 (自适应频谱+交互卷积) 🏛 arXiv 2024

MSGNet: Multi-Scale Inter-Series Correlations

🎯 多变量时序预测 (多尺度序列关联) 🏛 AAAI 2024

PatchTST: Time Series is Worth 64 Words

🎯 NLP时序预测 (补丁时序预测) 🏛 ICLR 2023

图像分割 (10个)

UACANet: Uncertainty Augmented Context Attention

🎯 息肉分割 (不确定性增强上下文注意力) 🏛 MICCAI 2021

SFFNet: Wavelet Spatial-Frequency Fusion Network

🎯 遥感分割 (小波空频融合) 🏛 arXiv 2024

DyCON: Dynamic Uncertainty-aware Consistency

🎯 半监督医学分割 (动态不确定性+对比学习) 🏛 CVPR 2025

Iris: In-Context Learning for Medical Segmentation

🎯 通用医学分割 (上下文参考引导/解耦架构) 🏛 CVPR 2025

ConText: In-Context Learning for Text Removal

🎯 文本移除+分割 (上下文学习驱动) 🏛 ICML 2025

RefLDMSeg: In-Context Segmentation via Latent DM

🎯 上下文分割 (潜在扩散模型) 🏛 AAAI 2025

HybridGL: Global-Local Representation + Spatial Guidance

🎯 零样本指代图像分割 🏛 CVPR 2025

TBConvL-Net: Hybrid CNN-Transformer Architecture

🎯 医学图像分割 (鲁棒混合架构) 🏛 arXiv 2024

HResFormer: Hybrid Residual Transformer

🎯 3D医学图像分割 (混合残差Transformer) 🏛 arXiv 2024

ScribFormer: CNN+Transformer for Scribble Segmentation

🎯 涂鸦监督医学分割 🏛 arXiv 2024

SAM系列 (14个)

ULSAM: Ultra-Lightweight Subspace Attention

🎯 轻量级CV (空间注意力) 🏛 WACV 2020

CSAM: Cross-Slice Attention Module

🎯 3D医学图像分割/CV通用 (交叉切片注意力) 🏛 WACV 2024

SAM2-LOVE: SAM2 in Language-Aided AV Scenes

🎯 视听场景分割 (多模态融合+Token传播) 🏛 CVPR 2025

CRISP-SAM2: Cross-Modal + Semantic Prompting

🎯 多器官分割 (跨模态交互+语义提示) 🏛 ACM MM 2025

SAM2-SGP: Support-Set Guided Prompting

🎯 医学分割 (支持集引导提示+伪掩码注意力) 🏛 arXiv 2025

SAMba-UNet: SAM2 + Mamba in UNet

🎯 心脏MRI分割 (SAM2+Mamba+UNet异构融合) 🏛 arXiv 2025

SAM2-UNet: SAM2 as Strong Encoder

🎯 自然+医学图像分割 (Hiera骨干+U-Net解码器) 🏛 ICCVW 2025

SAM2-UNeXT: High-Resolution Baseline

🎯 下游分割任务 (SAM2+DINOv2双分辨率) 🏛 arXiv 2025

MedSAM2: Segment Anything in 3D Medical

🎯 3D医学图像+视频分割 🏛 arXiv 2025

EfficientSAM: High-Resolution Generation + Perception

🎯 SAM轻量化 (高效基础视觉模型) 🏛 IEEE TPAMI 2024

RepViT-SAM: Real-Time Segment Anything

🎯 实时SAM (RepViT加速) 🏛 arXiv 2023

TinySAM: Efficient Segment Anything (USTC+Huawei)

🎯 高效SAM (突破分割极限) 🏛 AAAI 2024

EMA: Efficient Multi-Scale Attention (ESAM)

🎯 CV 2D通用 (增强边缘信息) 🏛 arXiv 2023

DSAM: Temporal-Spatial Brain Network Dynamics

🎯 图像恢复 (注意力模块) 🏛 arXiv 2024

目标检测 (7个)

FFCA-YOLO: Small Object Detection in RS

🎯 遥感小目标检测 (YOLO轻量级) 🏛 TGRS 2024

MAGNet: Multi-scale Awareness + Global fusion

🎯 RGB-D显著性目标检测 🏛 KBS 2024

RemoteDet-Mamba: Hybrid Mamba-CNN for RS Detection

🎯 遥感多模态目标检测 (Mamba-CNN混合) 🏛 arXiv 2024

U-DECN: Underwater Object Detection ConvNet

🎯 水下目标检测 (端到端去噪训练) 🏛 arXiv 2024

Low-light Object Detection

🎯 低光目标检测 🏛 arXiv 2024

DRPCA-Net: Robust PCA for Infrared Small Target

🎯 红外小目标检测 (鲁棒PCA) 🏛 arXiv 2025

Cross-view Representation for IR Small Target

🎯 红外小目标检测 (跨视角表征) 🏛 arXiv 2025

AI+医学 (7个)

CLEEGN: Plug-and-Play EEG Reconstruction

🎯 自动脑电图信号重建 (即插即用CNN) 🏛 arXiv 2022

GLSA: Global-Local Self-Attention (GLSANet)

🎯 医学图像分割 (全局-局部空间聚合) 🏛 PRCV 2023

SvANet: Scale-variant Attention Network

🎯 小型医学对象分割 🏛 arXiv 2024

EMCAD: Efficient Multi-scale Conv Attention Decoding

🎯 医学图像分割 (高效多尺度注意力解码) 🏛 CVPR 2024

DEFN: Dual-Encoder Fourier Group Harmonics

🎯 3D医学分割+重建 (模糊边界) 🏛 arXiv 2023

MASAG: Multi-scale Adaptive Spatial Attention Gate

🎯 医学图像分割 (多尺度自适应注意力门控) 🏛 BMVC 2024

Vision-LSTM: xLSTM as Generic Vision Backbone

🎯 医学图像分割 (xLSTM视觉主干) 🏛 arXiv 2024

CV所有任务 (3个)

RFAConv: Spatial Attention + Standard Convolution

🎯 CV通用 (分类/检测/分割) 🏛 arXiv 2023

SPABlock: Salient Positions based Attention

🎯 CV通用 (显著位置选择/非卷积非注意力) 🏛 arXiv 2021

SwiftFormer: Efficient Additive Attention

🎯 CV通用 (轻量高效编码器) 🏛 ICCV 2023

CV二维任务 (4个)

CoordGate: Spatially-Varying Convolutions

🎯 CV 2D通用 (动态权重调整/非卷积非注意力) 🏛 arXiv 2024

CAN: Context-Aware Module for Crowd Counting

🎯 CV 2D通用 (上下文感知/人群计数) 🏛 CVPR 2019

SSPCAB: Self-Supervised Predictive Conv Attentive Block

🎯 CV 2D通用 (异常检测/图像视频) 🏛 arXiv 2021

DynamicFilter: Dynamic Frequency Filtering

🎯 CV 2D通用 (频域滤波/动态权重) 🏛 NeurIPS 2016

图像超分 (4个)

FMB: Functional Manipulation Benchmark

🎯 轻量级即插即用超分模块 🏛 arXiv 2024

SAFMN: Spatially-Adaptive Feature Modulation

🎯 高效图像超分 🏛 ICCV 2023

ELAN: Efficient Long-Range Attention Network

🎯 图像超分 (高效长程注意力ELAB) 🏛 ECCV 2022

DAT: Dual Aggregation Transformer

🎯 图像超分 🏛 ICCV 2023

点云 (7个)

AdaptConv: Adaptive Graph Convolution for PC

🎯 点云分类+分割 (自适应图卷积) 🏛 ICCV 2021

GeoConv: Geodesic Guided Convolution

🎯 点云/人脸AU识别 (测地线引导卷积) 🏛 arXiv 2020

PnP-3D: Plug-and-Play for 3D Point Clouds

🎯 点云增强 (即插即用) 🏛 ICCV 2021

Point-NN: Non-parametric Point Cloud Analysis

🎯 点云分析 (非参数网络) 🏛 CVPR 2023

PF-Net: Point Fractal Network

🎯 点云补全 🏛 CVPR 2020

ISL: Intra-region Structure Learning (PRA-Net)

🎯 点云分析 (区域内结构学习) 🏛 arXiv 2021

KPConv: Kernel Point Convolution

🎯 点云特征提取 (灵活可变形点卷积) 🏛 ICCV 2019

视频预测 (1个)

SimVP: Simpler yet Better Video Prediction

🎯 视频预测 (简化高效) 🏛 CVPR 2022

3D任务 (4个)

PoseBERT: Generic Transformer for 3D Human

🎯 3D人体建模 (3D任务通用) 🏛 TPAMI 2023

GKONet: Geometric Knowledge 2D-to-3D Pose

🎯 3D人体姿态估计 (高维先验几何特征) 🏛 IEEE TCSVT 2023

Deformable LKA: Large Kernel Attention

🎯 3D视觉 (可变形大核注意力) 🏛 WACV 2024

MoE3D: Mixture-of-Experts for 3D Reconstruction

🎯 3D重建 (像素级深度边界锐度) 🏛 arXiv 2026

NLP (1个)

CorNet: Label Correlation Learning

🎯 NLP通用 (即插即用标签相关性学习) 🏛 IEEE Access 2019

语音识别 (1个)

FAdam: Natural Gradient Optimizer

🎯 语音/NLP/CV通用 (即插即用优化器) 🏛 arXiv 2024

人体姿态估计 (1个)

SmoothNet: Plug-and-Play Pose Refinement

🎯 2D/3D人体姿态估计 (姿态精炼) 🏛 ECCV 2022

Transformer/Unet专用 (1个)

DA_Block: Dual Attention (DANet)

🎯 场景分割 (可缝合在Transformer或UNet) 🏛 CVPR 2019

图像恢复 (3个)

NAF: Simple Baselines (NAFNet)

🎯 图像恢复 (即插即用NAF模块) 🏛 ECCV 2022

Histoformer: Histogram Transformer

🎯 恶劣天气图像恢复 🏛 arXiv 2024

AST: Adaptive Sparse Transformer

🎯 图像恢复 (自适应稀疏Transformer) 🏛 CVPR 2024

语义分割 (2个)

CGRSeg: Context-Guided Spatial Feature Reconstruction

🎯 语义分割 (RCM模块+DPG头) 🏛 ECCV 2024

CFBConv: Semantic Info CNN Conv (SCTNet)

🎯 实时语义分割 (CFBConv即插即用卷积) 🏛 AAAI 2024

图像增强 (1个)

FARM: Multi-Scale Feature Alignment (Burstormer)

🎯 图像增强/去噪/暗光/恢复/遥感 (多尺度对齐) 🏛 CVPR 2023

图像生成 (1个)

SeD: Semantic-Aware Discriminator

🎯 图像生成/超分 (GAN语义感知判别器) 🏛 CVPR 2024