cs.CV（2026-09-15）

📊 共 36 篇论文 | 🔗 5 篇有代码

🎯 兴趣领域导航

支柱二：RL算法与架构 (RL & Architecture) (15 🔗3) 支柱九：具身大模型 (Embodied Foundation Models) (10 🔗1) 支柱三：空间感知与语义 (Perception & Semantics) (5) 支柱一：机器人控制 (Robot Control) (3 🔗1) 支柱六：视频提取与匹配 (Video Extraction) (3)

🔬 支柱二：RL算法与架构 (RL & Architecture) (15 篇)

#题目一句话要点标签🔗⭐
1 SAVOR: Self-Aware Visual Grounding via Confidence-Calibrated Reinforcement Learning for Multimodal Hallucination Mitigation 提出SAVOR以解决多模态幻觉问题 reinforcement learning DPO large language model
2 sensVLA: Spatially-Grounded Vision-Language-Action Model for Autonomous Wheel Loader 提出sensVLA以解决自主轮式装载机控制问题 flow matching egocentric egocentric vision
3 HLC-GS: Risk-Map-Guided Height-Layer Consistency Gaussian Splatting for DSM Reconstruction from Optical Satellite Imagery 提出HLC-GS以解决DSM重建中的高度层混合问题 MAE 3D gaussian splatting 3DGS
4 Bi-FlowGS: Bridging Generative View Completion and Gaussian Geometry through Bidirectional Flow Co-Refinement 提出Bi-FlowGS以解决稀疏视图3D重建中的几何欺骗问题 distillation 3D gaussian splatting 3DGS
5 Unifying Semantic Priors and High-Frequency Traces: Enhancing V-JEPA with Mixture-of-Experts for Robust Synthetic Image Forensics 提出MoE-JEPA以解决深度伪造图像检测问题 world model world models JEPA
6 Hub-Spectral Activation of Latent Multimodal Knowledge 提出Hub-Spectral Activation以解决多模态知识激活问题 representation learning multimodal ✅
7 SPEAR NeXT Causal Latent Forecasting Across Multiple Horizons for Spectral Temporal Earth Representation Learning 提出SPEAR NeXT以解决地球观测中的时序预测问题 representation learning foundation model multimodal
8 RegRet: Enhancing Region-Level Retrieval in Large Multimodal Models 提出RegRet以解决大规模多模态模型中的区域级检索问题 contrastive learning multimodal
9 Efficient Quantization-Aware Distillation with Cross-Modal Alignment for Edge Vision-Language Models 提出统一框架以提高边缘视觉语言模型的量化蒸馏效率 contrastive learning distillation open-vocabulary
10 MedPCFM-TED: One-Step Point Cloud Flow Matching for Implant Generation via Teacher-Guided Endpoint Distillation 提出Teacher-Guided Endpoint Distillation以加速颅骨植入物生成 flow matching distillation
11 Can Knowledge Transfer Parameters Be Learned? LePoKet for Efficient Robotic Vision 提出LePoKet以解决机器人视觉中的知识转移问题 distillation optical flow motion estimation
12 MAETrack: Unleashing the Potential of Pretrained Geometric Priors for 3D Single Object Tracking 提出MAETrack以解决3D单目标跟踪中的预训练几何先验转移问题 representation learning masked autoencoder MAE
13 Hyper-RED: Scalable Event Pre-training via Semantic Hypergraph Distillation 提出Hyper-RED以解决事件表示学习中的数据稀缺问题 representation learning distillation ✅
14 Which Pretext Task Transfers? Self-Supervised Pretraining Objectives for Lung Ultrasound 比较三种自监督预训练目标以优化肺部超声图像分析 JEPA Joint-Embedding Predictive Architecture joint-embedding predictive architecture ✅
15 Channel-Wise and Token-Aware Post-Training Quantization for Visual State Space Duality 提出CTOAC方法以解决VSSD低比特量化问题 Mamba SSM state space model

🔬 支柱九：具身大模型 (Embodied Foundation Models) (10 篇)

#题目一句话要点标签🔗⭐
16 Semantic-Spatial Agreement Verification for Mitigating Object Hallucination in Multimodal Large Language Models 提出语义-空间一致性验证以解决多模态大语言模型的对象幻觉问题 large language model multimodal
17 What Do Hallucinations Reveal About Multimodal Reasoning? Diagnosing Visual Grounding Failures via Contrastive Decoding Probes 提出SAFE框架以诊断多模态推理中的视觉幻觉问题 multimodal visual grounding ✅
18 A Vision-Language Foundation Model for Precise and Comprehensive Brain Tumor Diagnosis from Preoperative Multimodal Data 提出BrainVLM以解决脑肿瘤诊断中的不确定性问题 foundation model multimodal
19 A multimodal large language model for evidence-based autism spectrum disorder screening 提出ASDchat以解决自闭症谱系障碍早期筛查问题 large language model multimodal
20 Multimodal Cultural Heritage Architectural Style Classification for Residential Buildings in the UAE Based on CLIP Embeddings and SVM 提出多模态框架以解决阿联酋住宅建筑风格分类问题 multimodal
21 Video-HolmesV2: Can MLLMs Reason with Spatio-Temporal Audio-Visual Evidence in Long Videos? 提出Video-HolmesV2以解决长视频中的多模态推理问题 large language model multimodal
22 ResLRP: The Role of Residual Cancellation in Attribution Instability in Vision Transformers 提出ResLRP以解决ViTs中的归因不稳定性问题 multimodal
23 High-Fidelity Video Quality Assessment with VQA-Specific Saliency 提出HFVQA框架以解决视频质量评估中的计算效率问题 foundation model
24 Multi-modal Knowledge Preserving Adapter for Embedding Backward Compatibility 提出多模态知识保留适配器以解决嵌入模型向后兼容性问题 large language model
25 ViD: Vision-Dominant Gender Bias Mitigation for Large Vision-Language Models 提出ViD框架以解决大型视觉语言模型中的性别偏见问题 visual grounding

🔬 支柱三：空间感知与语义 (Perception & Semantics) (5 篇)

#题目一句话要点标签🔗⭐
26 PanoGS-SLAM: Panoramic 3D Gaussian Splatting SLAM 提出PanoGS-SLAM以解决动态环境下的稠密SLAM问题 3D gaussian splatting 3DGS gaussian splatting
27 Optical-Flow Wingbeat Counting in MuJoCo: A Comparison of Convolutional, Spiking, and Attention-Based Temporal Models 提出光流翼击计数方法以解决无人机监测问题 optical flow
28 DecoGS: Adaptive Static-Dynamic Decoupling of 3D Gaussians for Free-Viewpoint Video Streaming 提出DecoGS以解决3D视频流中的静态动态分离问题 3D reconstruction spatiotemporal
29 Evaluating Mesh Reconstruction Methods for Crop Phenotyping 评估作物表型的网格重建方法以解决远程监测问题 3D reconstruction
30 G3AR: Graph-Guided Neural Visual Geometry for Scalable Multi-Sequence Aerial Registration 提出G3AR框架以解决多序列航拍图像配准问题 VGGT

🔬 支柱一：机器人控制 (Robot Control) (3 篇)

#题目一句话要点标签🔗⭐
31 MEgoVista: Multi-view Ego-aware Motion Estimation for Metric 4D Hands and Head in the Wild 提出MEgoVista以解决无约束环境下手部和头部运动估计问题 manipulation egocentric motion reconstruction
32 GeoLAM: Learning Geometry-Grounded Latent Actions from Unlabeled Human Videos 提出GeoLAM以从无标签人类视频中学习几何基础潜在动作 manipulation
33 PhysStream: Streaming Physics-Grounded Video Generation with Structured Scene Memory and Fine-Grained Motion Control 提出PhysStream以解决动态场景细粒度控制问题 manipulation ✅

🔬 支柱六：视频提取与匹配 (Video Extraction) (3 篇)

#题目一句话要点标签🔗⭐
34 EventEgoHands++: Event-based Egocentric 3D Hand Mesh Reconstruction with Real Dataset 提出EventEgoHands++以解决低光环境下的3D手部重建问题 egocentric hand reconstruction spatial relationship
35 FROD: Feature Matching Residual Denoising Oracle Bone Decipher 提出FROD以解决Oracle骨文解读中的噪声与结构问题 feature matching
36 EgoPathBench: Evaluating Zero-Shot Egocentric Waypoint Decision-Making in Vision-Language Models 提出EgoPathBench以解决零-shot自我中心路径决策问题 egocentric

⬅️ 返回 cs.CV 首页 · 🏠 返回主页