cs.CV（2025-05-11）

📊 共 16 篇论文 | 🔗 4 篇有代码

🎯 兴趣领域导航

支柱九：具身大模型 (Embodied Foundation Models) (7 🔗2) 支柱二：RL算法与架构 (RL & Architecture) (4 🔗2) 支柱三：空间感知与语义 (Perception & Semantics) (3) 支柱七：动作重定向 (Motion Retargeting) (1) 支柱四：生成式动作 (Generative Motion) (1)

🔬 支柱九：具身大模型 (Embodied Foundation Models) (7 篇)

#题目一句话要点标签🔗⭐
1 Hallucination-Aware Multimodal Benchmark for Gastrointestinal Image Analysis with Large Vision-Language Models 提出Gut-VLM数据集，并采用幻觉感知微调方法提升VLM在胃肠道图像分析中的可靠性。 multimodal ✅
2 Seed1.5-VL Technical Report 提出Seed1.5-VL，用于提升通用多模态理解和推理能力 foundation model multimodal
3 Building a Human-Verified Clinical Reasoning Dataset via a Human LLM Hybrid Pipeline for Trustworthy Medical AI 构建基于人机混合流程的临床推理数据集，提升医疗AI的可信度 large language model chain-of-thought
4 Visual Instruction Tuning with Chain of Region-of-Interest 提出基于感兴趣区域链的视觉指令调优方法CoRoI，提升高分辨率图像多模态大模型的效率。 large language model multimodal
5 DeepSORT-Driven Visual Tracking Approach for Gesture Recognition in Interactive Systems 利用DeepSORT的视觉跟踪方法用于交互系统中手势识别 multimodal
6 DAPE: Dual-Stage Parameter-Efficient Fine-Tuning for Consistent Video Editing with Diffusion Models DAPE：双阶段参数高效微调框架，用于扩散模型视频一致性编辑 multimodal
7 Efficient and Robust Multidimensional Attention in Remote Physiological Sensing through Target Signal Constrained Factorization 提出TSFM约束的多维注意力机制，提升远程生理信号感知的跨域泛化能力 multimodal ✅

🔬 支柱二：RL算法与架构 (RL & Architecture) (4 篇)

#题目一句话要点标签🔗⭐
8 Multimodal Fake News Detection: MFND Dataset and Shallow-Deep Multitask Learning 提出MFND数据集和浅层-深层多任务学习模型SDML，用于检测和定位多模态假新闻。 contrastive learning distillation multimodal ✅
9 A Vision-Language Foundation Model for Leaf Disease Identification 提出SCOLD，一种用于叶片病害识别的视觉-语言基础模型。 contrastive learning foundation model ✅
10 Joint Low-level and High-level Textual Representation Learning with Multiple Masking Strategies 提出多重掩码策略MMS，联合学习文本识别的低级纹理和高级语义表示。 representation learning masked autoencoder MAE
11 Enhancing Monocular Height Estimation via Sparse LiDAR-Guided Correction 提出稀疏LiDAR引导的校正方法，提升单目高程估计精度与鲁棒性 MAE depth estimation monocular depth

🔬 支柱三：空间感知与语义 (Perception & Semantics) (3 篇)

#题目一句话要点标签🔗⭐
12 MELLM: A Flow-Guided Large Language Model for Micro-Expression Understanding 提出MELLM，利用光流引导的大语言模型进行微表情理解 optical flow large language model multimodal
13 NeuGen: Amplifying the 'Neural' in Neural Radiance Fields for Domain Generalization 提出NeuGen，增强NeRF的泛化能力，提升跨域场景渲染质量 NeRF neural radiance field
14 Technical Report for ICRA 2025 GOOSE 2D Semantic Segmentation Challenge: Leveraging Color Shift Correction, RoPE-Swin Backbone, and Quantile-based Label Denoising Strategy for Robust Outdoor Scene Understanding 针对户外场景语义分割，提出颜色校正、RoPE-Swin和分位数去噪的稳健框架 scene understanding

🔬 支柱七：动作重定向 (Motion Retargeting) (1 篇)

#题目一句话要点标签🔗⭐
15 Differentiable NMS via Sinkhorn Matching for End-to-End Fabric Defect Detection 提出基于Sinkhorn匹配的可微NMS，用于端到端织物缺陷检测 spatial relationship

🔬 支柱四：生成式动作 (Generative Motion) (1 篇)

#题目一句话要点标签🔗⭐
16 Privacy of Groups in Dense Street Imagery 揭示城市街景图像中群体隐私泄露风险，提出针对性隐私保护建议 penetration

⬅️ 返回 cs.CV 首页 · 🏠 返回主页